"git@developer.sourcefind.cn:wangsen/paddle_dbnet.git" did not exist on "8235eece1e11148c722f2409a5af205fe064461f"
Commit 4c4cad8b authored by Leif's avatar Leif
Browse files

Add PPOCRLabel

parent 569deedc
# ex: set ts=8 noet:
all: qt5 test
test: testpy3
testpy2:
python -m unittest discover tests
testpy3:
python3 -m unittest discover tests
qt4: qt4py2
qt5: qt5py3
qt4py2:
pyrcc4 -py2 -o libs/resources.py resources.qrc
qt4py3:
pyrcc4 -py3 -o libs/resources.py resources.qrc
qt5py3:
pyrcc5 -o libs/resources.py resources.qrc
clean:
rm -rf ~/.labelImgSettings.pkl *.pyc dist labelImg.egg-info __pycache__ build
pip_upload:
python3 setup.py upload
long_description:
restview --long-description
.PHONY: all
# Copyright (c) <2015-Present> Tzutalin
# Copyright (C) 2013 MIT, Computer Science and Artificial Intelligence Laboratory. Bryan Russell, Antonio Torralba,
# William T. Freeman. Permission is hereby granted, free of charge, to any person obtaining a copy of this software and
# associated documentation files (the "Software"), to deal in the Software without restriction, including without
# limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the
# Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
# The above copyright notice and this permission notice shall be included in all copies or substantial portions of
# the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT
# NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT
# SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF
# CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
# THE SOFTWARE.
#!/usr/bin/env python
# -*- coding: utf-8 -*-
# pyrcc5 -o libs/resources.py resources.qrc
import argparse
import ast
import codecs
import os.path
import platform
import subprocess
import sys
from functools import partial
from collections import defaultdict
import json
__dir__ = os.path.dirname(os.path.abspath(__file__))
sys.path.append(__dir__)
sys.path.append(os.path.abspath(os.path.join(__dir__, '../..')))
sys.path.append("..")
from paddleocr import PaddleOCR
try:
from PyQt5 import QtCore, QtGui, QtWidgets
from PyQt5.QtGui import *
from PyQt5.QtCore import *
from PyQt5.QtWidgets import *
except ImportError:
# needed for py3+qt4
# Ref:
# http://pyqt.sourceforge.net/Docs/PyQt4/incompatible_apis.html
# http://stackoverflow.com/questions/21217399/pyqt4-qtcore-qvariant-object-instead-of-a-string
if sys.version_info.major >= 3:
import sip
sip.setapi('QVariant', 2)
from PyQt4.QtGui import *
from PyQt4.QtCore import *
from combobox import ComboBox
from libs.constants import *
from libs.utils import *
from libs.settings import Settings
from libs.shape import Shape, DEFAULT_LINE_COLOR, DEFAULT_FILL_COLOR
from libs.stringBundle import StringBundle
from libs.canvas import Canvas
from libs.zoomWidget import ZoomWidget
from libs.autoDialog import AutoDialog
from libs.labelDialog import LabelDialog
from libs.colorDialog import ColorDialog
from libs.labelFile import LabelFile, LabelFileError
from libs.toolBar import ToolBar
from libs.ustr import ustr
from libs.hashableQListWidgetItem import HashableQListWidgetItem
__appname__ = 'PPOCRLabel'
class WindowMixin(object):
def menu(self, title, actions=None):
menu = self.menuBar().addMenu(title)
if actions:
addActions(menu, actions)
return menu
def toolbar(self, title, actions=None):
toolbar = ToolBar(title)
toolbar.setObjectName(u'%sToolBar' % title)
# toolbar.setOrientation(Qt.Vertical)
toolbar.setToolButtonStyle(Qt.ToolButtonTextUnderIcon)
if actions:
addActions(toolbar, actions)
self.addToolBar(Qt.LeftToolBarArea, toolbar)
return toolbar
class MainWindow(QMainWindow, WindowMixin):
FIT_WINDOW, FIT_WIDTH, MANUAL_ZOOM = list(range(3))
def __init__(self, lang="ch", defaultFilename=None, defaultPrefdefClassFile=None, defaultSaveDir=None):
super(MainWindow, self).__init__()
self.setWindowTitle(__appname__)
# Load setting in the main thread
self.settings = Settings()
self.settings.load()
settings = self.settings
self.lang = lang
# Load string bundle for i18n
if lang not in ['ch', 'en']:
lang = 'en'
self.stringBundle = StringBundle.getBundle(localeStr='zh-CN' if lang=='ch' else 'en') # 'en'
getStr = lambda strId: self.stringBundle.getString(strId)
self.defaultSaveDir = defaultSaveDir
self.ocr = PaddleOCR(use_pdserving=False, use_angle_cls=True, det=True, cls=True, use_gpu=True, lang=lang)
if os.path.exists('./data/paddle.png'):
result = self.ocr.ocr('./data/paddle.png', cls=True, det=True)
# For loading all image under a directory
self.mImgList = []
self.mImgList5 = []
self.dirname = None
self.labelHist = []
self.lastOpenDir = None
self.result_dic = []
self.changeFileFolder = False
self.haveAutoReced = False
self.labelFile = None
self.currIndex = 0
# Whether we need to save or not.
self.dirty = False
self._noSelectionSlot = False
self._beginner = True
self.screencastViewer = self.getAvailableScreencastViewer()
self.screencast = "https://github.com/PaddlePaddle/PaddleOCR"
# Load predefined classes to the list
self.loadPredefinedClasses(defaultPrefdefClassFile)
# Main widgets and related state.
self.labelDialog = LabelDialog(parent=self, listItem=self.labelHist)
self.autoDialog = AutoDialog(parent=self)
self.itemsToShapes = {}
self.shapesToItems = {}
self.itemsToShapesbox = {}
self.shapesToItemsbox = {}
self.prevLabelText = getStr('tempLabel')
self.model = 'paddle'
self.PPreader = None
self.autoSaveNum = 10
################# file list ###############
self.fileListWidget = QListWidget()
self.fileListWidget.itemClicked.connect(self.fileitemDoubleClicked)
self.fileListWidget.setIconSize(QSize(25, 25))
filelistLayout = QVBoxLayout()
filelistLayout.setContentsMargins(0, 0, 0, 0)
filelistLayout.addWidget(self.fileListWidget)
self.AutoRecognition = QToolButton()
self.AutoRecognition.setToolButtonStyle(Qt.ToolButtonTextBesideIcon)
self.AutoRecognition.setIcon(newIcon('Auto'))
autoRecLayout = QHBoxLayout()
autoRecLayout.setContentsMargins(0, 0, 0, 0)
autoRecLayout.addWidget(self.AutoRecognition)
autoRecContainer = QWidget()
autoRecContainer.setLayout(autoRecLayout)
filelistLayout.addWidget(autoRecContainer)
fileListContainer = QWidget()
fileListContainer.setLayout(filelistLayout)
self.filedock = QDockWidget(getStr('fileList'), self)
self.filedock.setObjectName(getStr('files'))
self.filedock.setWidget(fileListContainer)
self.addDockWidget(Qt.LeftDockWidgetArea, self.filedock)
######## Right area ##########
listLayout = QVBoxLayout()
listLayout.setContentsMargins(0, 0, 0, 0)
self.editButton = QToolButton()
self.reRecogButton = QToolButton()
self.reRecogButton.setIcon(newIcon('reRec', 30))
self.reRecogButton.setToolButtonStyle(Qt.ToolButtonTextBesideIcon)
self.newButton = QToolButton()
self.newButton.setToolButtonStyle(Qt.ToolButtonTextBesideIcon)
self.SaveButton = QToolButton()
self.SaveButton.setToolButtonStyle(Qt.ToolButtonTextBesideIcon)
self.DelButton = QToolButton()
self.DelButton.setToolButtonStyle(Qt.ToolButtonTextBesideIcon)
lefttoptoolbox = QHBoxLayout()
lefttoptoolbox.addWidget(self.newButton)
lefttoptoolbox.addWidget(self.reRecogButton)
lefttoptoolboxcontainer = QWidget()
lefttoptoolboxcontainer.setLayout(lefttoptoolbox)
listLayout.addWidget(lefttoptoolboxcontainer)
################## label list ####################
# Create and add a widget for showing current label items
self.labelList = QListWidget()
labelListContainer = QWidget()
labelListContainer.setLayout(listLayout)
self.labelList.itemActivated.connect(self.labelSelectionChanged)
self.labelList.itemSelectionChanged.connect(self.labelSelectionChanged)
self.labelList.itemDoubleClicked.connect(self.editLabel)
# Connect to itemChanged to detect checkbox changes.
self.labelList.itemChanged.connect(self.labelItemChanged)
self.labelListDock = QDockWidget(getStr('recognitionResult'),self)
self.labelListDock.setWidget(self.labelList)
self.labelListDock.setFeatures(QDockWidget.NoDockWidgetFeatures)
listLayout.addWidget(self.labelListDock)
################## detection box ####################
self.BoxList = QListWidget()
self.BoxList.itemActivated.connect(self.boxSelectionChanged)
self.BoxList.itemSelectionChanged.connect(self.boxSelectionChanged)
self.BoxList.itemDoubleClicked.connect(self.editBox)
# Connect to itemChanged to detect checkbox changes.
self.BoxList.itemChanged.connect(self.boxItemChanged)
self.BoxListDock = QDockWidget(getStr('detectionBoxposition'), self)
self.BoxListDock.setWidget(self.BoxList)
self.BoxListDock.setFeatures(QDockWidget.NoDockWidgetFeatures)
listLayout.addWidget(self.BoxListDock)
############ lower right area ############
leftbtmtoolbox = QHBoxLayout()
leftbtmtoolbox.addWidget(self.SaveButton)
leftbtmtoolbox.addWidget(self.DelButton)
leftbtmtoolboxcontainer = QWidget()
leftbtmtoolboxcontainer.setLayout(leftbtmtoolbox)
listLayout.addWidget(leftbtmtoolboxcontainer)
self.dock = QDockWidget(getStr('boxLabelText'), self)
self.dock.setObjectName(getStr('labels'))
self.dock.setWidget(labelListContainer)
########## zoom bar #########
self.imgsplider = QSlider(Qt.Horizontal)
self.imgsplider.valueChanged.connect(self.CanvasSizeChange)
self.imgsplider.setMinimum(-150)
self.imgsplider.setMaximum(150)
self.imgsplider.setSingleStep(1)
self.imgsplider.setTickPosition(QSlider.TicksBelow)
self.imgsplider.setTickInterval(1)
op = QGraphicsOpacityEffect()
op.setOpacity(0.2)
self.imgsplider.setGraphicsEffect(op)
# self.imgsplider.setAttribute(Qt.WA_TranslucentBackground)
self.imgsplider.setStyleSheet("background-color:transparent")
self.imgsliderDock = QDockWidget(getStr('ImageResize'), self)
self.imgsliderDock.setObjectName(getStr('IR'))
self.imgsliderDock.setWidget(self.imgsplider)
self.imgsliderDock.setFeatures(QDockWidget.DockWidgetFloatable)
# op = QGraphicsOpacityEffect()
# op.setOpacity(0.2)
# self.imgsliderDock.setGraphicsEffect(op)
self.imgsliderDock.setAttribute(Qt.WA_TranslucentBackground)
self.addDockWidget(Qt.RightDockWidgetArea, self.imgsliderDock)
self.zoomWidget = ZoomWidget()
self.colorDialog = ColorDialog(parent=self)
self.zoomWidgetValue = self.zoomWidget.value()
########## thumbnail #########
hlayout = QHBoxLayout()
m = (0, 0, 0, 0)
hlayout.setSpacing(0)
hlayout.setContentsMargins(*m)
self.preButton = QToolButton()
self.preButton.setIcon(newIcon("prev",40))
self.preButton.setIconSize(QSize(40, 100))
self.preButton.clicked.connect(self.openPrevImg)
self.preButton.setStyleSheet('border: none;')
self.iconlist = QListWidget()
self.iconlist.setViewMode(QListView.IconMode)
self.iconlist.setFlow(QListView.TopToBottom)
self.iconlist.setSpacing(10)
self.iconlist.setIconSize(QSize(50, 50))
self.iconlist.setMovement(False)
self.iconlist.setResizeMode(QListView.Adjust)
self.iconlist.itemClicked.connect(self.iconitemDoubleClicked)
self.iconlist.setStyleSheet("background-color:transparent; border: none;")
self.iconlist.setHorizontalScrollBarPolicy(Qt.ScrollBarAlwaysOff)
self.nextButton = QToolButton()
self.nextButton.setIcon(newIcon("next", 40))
self.nextButton.setIconSize(QSize(40, 100))
self.nextButton.setStyleSheet('border: none;')
self.nextButton.clicked.connect(self.openNextImg)
hlayout.addWidget(self.preButton)
hlayout.addWidget(self.iconlist)
hlayout.addWidget(self.nextButton)
# self.setLayout(hlayout)
iconListContainer = QWidget()
iconListContainer.setLayout(hlayout)
iconListContainer.setFixedHeight(100)
# iconListContainer.setFixedWidth(530)
# op = QGraphicsOpacityEffect()
# op.setOpacity(0.5)
# iconListContainer.setGraphicsEffect(op)
########### Canvas ###########
self.canvas = Canvas(parent=self)
self.canvas.zoomRequest.connect(self.zoomRequest)
self.canvas.setDrawingShapeToSquare(settings.get(SETTING_DRAW_SQUARE, False))
scroll = QScrollArea()
scroll.setWidget(self.canvas)
scroll.setWidgetResizable(True)
self.scrollBars = {
Qt.Vertical: scroll.verticalScrollBar(),
Qt.Horizontal: scroll.horizontalScrollBar()
}
self.scrollArea = scroll
self.canvas.scrollRequest.connect(self.scrollRequest)
self.canvas.newShape.connect(self.newShape)
self.canvas.shapeMoved.connect(self.updateBoxlist) # self.setDirty
self.canvas.selectionChanged.connect(self.shapeSelectionChanged)
self.canvas.drawingPolygon.connect(self.toggleDrawingSensitive)
centerLayout = QVBoxLayout()
centerLayout.setContentsMargins(0, 0, 0, 0)
centerLayout.addWidget(scroll)
#centerLayout.addWidget(self.icondock)
centerLayout.addWidget(iconListContainer,0,Qt.AlignCenter)
centercontainer = QWidget()
centercontainer.setLayout(centerLayout)
# self.scrolldock = QDockWidget('WorkSpace',self)
# self.scrolldock.setObjectName('WorkSpace')
# self.scrolldock.setWidget(centercontainer)
# self.scrolldock.setFeatures(QDockWidget.NoDockWidgetFeatures)
# orititle = self.scrolldock.titleBarWidget()
# tmpwidget = QWidget()
# self.scrolldock.setTitleBarWidget(tmpwidget)
# del orititle
self.setCentralWidget(centercontainer) #self.scrolldock
self.addDockWidget(Qt.RightDockWidgetArea, self.dock)
# self.filedock.setFeatures(QDockWidget.DockWidgetFloatable)
self.filedock.setFeatures(self.filedock.features() ^ QDockWidget.DockWidgetFloatable)
self.dockFeatures = QDockWidget.DockWidgetClosable | QDockWidget.DockWidgetFloatable
self.dock.setFeatures(self.dock.features() ^ self.dockFeatures)
self.filedock.setFeatures(QDockWidget.NoDockWidgetFeatures)
###### Actions #######
action = partial(newAction, self)
quit = action(getStr('quit'), self.close,
'Ctrl+Q', 'quit', getStr('quitApp'))
open = action(getStr('openFile'), self.openFile,
'Ctrl+O', 'open', getStr('openFileDetail'))
opendir = action(getStr('openDir'), self.openDirDialog,
'Ctrl+u', 'open', getStr('openDir'))
openNextImg = action(getStr('nextImg'), self.openNextImg,
'd', 'next', getStr('nextImgDetail'))
openPrevImg = action(getStr('prevImg'), self.openPrevImg,
'a', 'prev', getStr('prevImgDetail'))
verify = action(getStr('verifyImg'), self.verifyImg,
'space', 'verify', getStr('verifyImgDetail'))
save = action(getStr('save'), self.saveFile,
'Ctrl+S', 'save', getStr('saveDetail'), enabled=False)
alcm = action(getStr('choosemodel'), self.autolcm,
'Ctrl+M', 'next', getStr('tipchoosemodel'))
deleteImg = action(getStr('deleteImg'), self.deleteImg, 'Ctrl+D', 'close', getStr('deleteImgDetail'),
enabled=True)
resetAll = action(getStr('resetAll'), self.resetAll, None, 'resetall', getStr('resetAllDetail'))
color1 = action(getStr('boxLineColor'), self.chooseColor1,
'Ctrl+L', 'color_line', getStr('boxLineColorDetail'))
createMode = action(getStr('crtBox'), self.setCreateMode,
'w', 'new', getStr('crtBoxDetail'), enabled=False)
editMode = action('&Edit\nRectBox', self.setEditMode,
'Ctrl+J', 'edit', u'Move and edit Boxs', enabled=False)
create = action(getStr('crtBox'), self.createShape,
'w', 'new', getStr('crtBoxDetail'), enabled=False)
delete = action(getStr('delBox'), self.deleteSelectedShape,
'Delete', 'delete', getStr('delBoxDetail'), enabled=False)
copy = action(getStr('dupBox'), self.copySelectedShape,
'Ctrl+D', 'copy', getStr('dupBoxDetail'),
enabled=False)
hideAll = action(getStr('hideBox'), partial(self.togglePolygons, False),
'Ctrl+H', 'hide', getStr('hideAllBoxDetail'),
enabled=False)
showAll = action(getStr('showBox'), partial(self.togglePolygons, True),
'Ctrl+A', 'hide', getStr('showAllBoxDetail'),
enabled=False)
help = action(getStr('tutorial'), self.showTutorialDialog, None, 'help', getStr('tutorialDetail'))
showInfo = action(getStr('info'), self.showInfoDialog, None, 'help', getStr('info'))
showSteps = action(getStr('steps'), self.showStepsDialog, None, 'help', getStr('steps'))
zoom = QWidgetAction(self)
zoom.setDefaultWidget(self.zoomWidget)
self.zoomWidget.setWhatsThis(
u"Zoom in or out of the image. Also accessible with"
" %s and %s from the canvas." % (fmtShortcut("Ctrl+[-+]"),
fmtShortcut("Ctrl+Wheel")))
self.zoomWidget.setEnabled(False)
zoomIn = action(getStr('zoomin'), partial(self.addZoom, 10),
'Ctrl++', 'zoom-in', getStr('zoominDetail'), enabled=False)
zoomOut = action(getStr('zoomout'), partial(self.addZoom, -10),
'Ctrl+-', 'zoom-out', getStr('zoomoutDetail'), enabled=False)
zoomOrg = action(getStr('originalsize'), partial(self.setZoom, 100),
'Ctrl+=', 'zoom', getStr('originalsizeDetail'), enabled=False)
fitWindow = action(getStr('fitWin'), self.setFitWindow,
'Ctrl+F', 'fit-window', getStr('fitWinDetail'),
checkable=True, enabled=False)
fitWidth = action(getStr('fitWidth'), self.setFitWidth,
'Ctrl+Shift+F', 'fit-width', getStr('fitWidthDetail'),
checkable=True, enabled=False)
# Group zoom controls into a list for easier toggling.
zoomActions = (self.zoomWidget, zoomIn, zoomOut,
zoomOrg, fitWindow, fitWidth)
self.zoomMode = self.MANUAL_ZOOM
self.scalers = {
self.FIT_WINDOW: self.scaleFitWindow,
self.FIT_WIDTH: self.scaleFitWidth,
# Set to one to scale to 100% when loading files.
self.MANUAL_ZOOM: lambda: 1,
}
edit = action(getStr('editLabel'), self.editLabel,
'Ctrl+E', 'edit', getStr('editLabelDetail'),
enabled=False)
######## New actions #######
AutoRec = action(getStr('autoRecognition'), self.autoRecognition,
'Ctrl+Shift+A', 'Auto', getStr('autoRecognition'), enabled=False)
reRec = action(getStr('reRecognition'), self.reRecognition,
'Ctrl+Shift+R', 'reRec', getStr('reRecognition'), enabled=False)
createpoly = action(getStr('creatPolygon'), self.createPolygon,
'p', 'new', 'Creat Polygon', enabled=True)
saveRec = action(getStr('saveRec'), self.saveRecResult,
'', 'saveRec', getStr('saveRec'), enabled=False)
self.editButton.setDefaultAction(edit)
self.newButton.setDefaultAction(create)
self.DelButton.setDefaultAction(deleteImg)
self.SaveButton.setDefaultAction(save)
self.AutoRecognition.setDefaultAction(AutoRec)
self.reRecogButton.setDefaultAction(reRec)
# self.preButton.setDefaultAction(openPrevImg)
# self.nextButton.setDefaultAction(openNextImg)
############# Zoom layout ##############
zoomLayout = QHBoxLayout()
zoomLayout.addStretch()
self.zoominButton = QToolButton()
self.zoominButton.setToolButtonStyle(Qt.ToolButtonTextBesideIcon)
self.zoominButton.setDefaultAction(zoomIn)
self.zoomoutButton = QToolButton()
self.zoomoutButton.setToolButtonStyle(Qt.ToolButtonTextBesideIcon)
self.zoomoutButton.setDefaultAction(zoomOut)
self.zoomorgButton = QToolButton()
self.zoomorgButton.setToolButtonStyle(Qt.ToolButtonTextBesideIcon)
self.zoomorgButton.setDefaultAction(zoomOrg)
zoomLayout.addWidget(self.zoominButton)
zoomLayout.addWidget(self.zoomorgButton)
zoomLayout.addWidget(self.zoomoutButton)
zoomContainer = QWidget()
zoomContainer.setLayout(zoomLayout)
zoomContainer.setGeometry(0, 0, 30, 150)
shapeLineColor = action(getStr('shapeLineColor'), self.chshapeLineColor,
icon='color_line', tip=getStr('shapeLineColorDetail'),
enabled=False)
shapeFillColor = action(getStr('shapeFillColor'), self.chshapeFillColor,
icon='color', tip=getStr('shapeFillColorDetail'),
enabled=False)
# labels = self.dock.toggleViewAction()
# labels.setText(getStr('showHide'))
# labels.setShortcut('Ctrl+Shift+L')
# Label list context menu.
labelMenu = QMenu()
addActions(labelMenu, (edit, delete))
self.labelList.setContextMenuPolicy(Qt.CustomContextMenu)
self.labelList.customContextMenuRequested.connect(
self.popLabelListMenu)
# Draw squares/rectangles
self.drawSquaresOption = QAction(getStr('drawSquares'), self)
self.drawSquaresOption.setShortcut('Ctrl+Shift+R')
self.drawSquaresOption.setCheckable(True)
self.drawSquaresOption.setChecked(settings.get(SETTING_DRAW_SQUARE, False))
self.drawSquaresOption.triggered.connect(self.toogleDrawSquare)
# Store actions for further handling.
self.actions = struct(save=save, open=open, resetAll=resetAll, deleteImg=deleteImg,
lineColor=color1, create=create, delete=delete, edit=edit, copy=copy,
saveRec=saveRec,
createMode=createMode, editMode=editMode,
shapeLineColor=shapeLineColor, shapeFillColor=shapeFillColor,
zoom=zoom, zoomIn=zoomIn, zoomOut=zoomOut, zoomOrg=zoomOrg,
fitWindow=fitWindow, fitWidth=fitWidth,
zoomActions=zoomActions,
fileMenuActions=(
open, opendir, save, resetAll, quit),
beginner=(), advanced=(),
editMenu=(createpoly, edit, copy, delete,
None, color1, self.drawSquaresOption),
beginnerContext=(create, edit, copy, delete),
advancedContext=(createMode, editMode, edit, copy,
delete, shapeLineColor, shapeFillColor),
onLoadActive=(
create, createMode, editMode),
onShapesPresent=(hideAll, showAll))
# menus
self.menus = struct(
file=self.menu('&'+getStr('mfile')),
edit=self.menu('&'+getStr('medit')),
view=self.menu('&'+getStr('mview')),
autolabel=self.menu('&PaddleOCR'),
help=self.menu('&'+getStr('mhelp')),
recentFiles=QMenu('Open &Recent'),
labelList=labelMenu)
# Sync single class mode from PR#106
self.singleClassMode = QAction(getStr('singleClsMode'), self)
self.singleClassMode.setShortcut("Ctrl+Shift+S")
self.singleClassMode.setCheckable(True)
self.singleClassMode.setChecked(settings.get(SETTING_SINGLE_CLASS, False))
self.lastLabel = None
# Add option to enable/disable labels being displayed at the top of bounding boxes
self.displayLabelOption = QAction(getStr('displayLabel'), self)
self.displayLabelOption.setShortcut("Ctrl+Shift+P")
self.displayLabelOption.setCheckable(True)
self.displayLabelOption.setChecked(settings.get(SETTING_PAINT_LABEL, False))
self.displayLabelOption.triggered.connect(self.togglePaintLabelsOption)
addActions(self.menus.file,
(opendir, None, save, resetAll, deleteImg, quit))
addActions(self.menus.help, (showSteps, showInfo))
addActions(self.menus.view, (
self.displayLabelOption, # labels,
None,
hideAll, showAll, None,
zoomIn, zoomOut, zoomOrg, None,
fitWindow, fitWidth))
addActions(self.menus.autolabel, (alcm, saveRec, None, help)) #
self.menus.file.aboutToShow.connect(self.updateFileMenu)
# Custom context menu for the canvas widget:
addActions(self.canvas.menus[0], self.actions.beginnerContext)
addActions(self.canvas.menus[1], (
action('&Copy here', self.copyShape),
action('&Move here', self.moveShape)))
# self.tools = self.toolbar('Tools')
self.actions.beginner = (
open, opendir, openNextImg, openPrevImg, verify, save, None, create, copy, delete, None,
zoomIn, zoom, zoomOut, fitWindow, fitWidth)
self.actions.advanced = (
open, opendir, openNextImg, openPrevImg, save, None,
createMode, editMode, None,
hideAll, showAll)
self.statusBar().showMessage('%s started.' % __appname__)
self.statusBar().show()
# Application state.
self.image = QImage()
self.filePath = ustr(defaultFilename)
self.lastOpenDir = None
self.recentFiles = []
self.maxRecent = 7
self.lineColor = None
self.fillColor = None
self.zoom_level = 100
self.fit_window = False
# Add Chris
self.difficult = False
## Fix the compatible issue for qt4 and qt5. Convert the QStringList to python list
if settings.get(SETTING_RECENT_FILES):
if have_qstring():
recentFileQStringList = settings.get(SETTING_RECENT_FILES)
self.recentFiles = [ustr(i) for i in recentFileQStringList]
else:
self.recentFiles = recentFileQStringList = settings.get(SETTING_RECENT_FILES)
size = settings.get(SETTING_WIN_SIZE, QSize(1200, 800))
position = QPoint(0, 0)
saved_position = settings.get(SETTING_WIN_POSE, position)
# Fix the multiple monitors issue
for i in range(QApplication.desktop().screenCount()):
if QApplication.desktop().availableGeometry(i).contains(saved_position):
position = saved_position
break
self.resize(size)
self.move(position)
saveDir = ustr(settings.get(SETTING_SAVE_DIR, None))
self.lastOpenDir = ustr(settings.get(SETTING_LAST_OPEN_DIR, None))
self.restoreState(settings.get(SETTING_WIN_STATE, QByteArray()))
Shape.line_color = self.lineColor = QColor(settings.get(SETTING_LINE_COLOR, DEFAULT_LINE_COLOR))
Shape.fill_color = self.fillColor = QColor(settings.get(SETTING_FILL_COLOR, DEFAULT_FILL_COLOR))
self.canvas.setDrawingColor(self.lineColor)
# Add chris
Shape.difficult = self.difficult
# ADD:
# Populate the File menu dynamically.
self.updateFileMenu()
# Since loading the file may take some time, make sure it runs in the background.
if self.filePath and os.path.isdir(self.filePath):
self.queueEvent(partial(self.importDirImages, self.filePath or ""))
elif self.filePath:
self.queueEvent(partial(self.loadFile, self.filePath or ""))
# Callbacks:
self.zoomWidget.valueChanged.connect(self.paintCanvas)
self.populateModeActions()
# Display cursor coordinates at the right of status bar
self.labelCoordinates = QLabel('')
self.statusBar().addPermanentWidget(self.labelCoordinates)
# Open Dir if deafult file
if self.filePath and os.path.isdir(self.filePath):
self.openDirDialog(dirpath=self.filePath, silent=True)
def keyReleaseEvent(self, event):
if event.key() == Qt.Key_Control:
self.canvas.setDrawingShapeToSquare(False)
def keyPressEvent(self, event):
if event.key() == Qt.Key_Control:
# Draw rectangle if Ctrl is pressed
self.canvas.setDrawingShapeToSquare(True)
def noShapes(self):
return not self.itemsToShapes
def populateModeActions(self):
self.canvas.menus[0].clear()
addActions(self.canvas.menus[0], self.actions.beginnerContext)
self.menus.edit.clear()
actions = (self.actions.create,) # if self.beginner() else (self.actions.createMode, self.actions.editMode)
addActions(self.menus.edit, actions + self.actions.editMenu)
def setDirty(self):
self.dirty = True
self.actions.save.setEnabled(True)
def setClean(self):
self.dirty = False
self.actions.save.setEnabled(False)
self.actions.create.setEnabled(True)
def toggleActions(self, value=True):
"""Enable/Disable widgets which depend on an opened image."""
for z in self.actions.zoomActions:
z.setEnabled(value)
for action in self.actions.onLoadActive:
action.setEnabled(value)
def queueEvent(self, function):
QTimer.singleShot(0, function)
def status(self, message, delay=5000):
self.statusBar().showMessage(message, delay)
def resetState(self):
self.itemsToShapes.clear()
self.shapesToItems.clear()
self.itemsToShapesbox.clear() # ADD
self.shapesToItemsbox.clear()
self.labelList.clear()
self.BoxList.clear()
self.filePath = None
self.imageData = None
self.labelFile = None
self.canvas.resetState()
self.labelCoordinates.clear()
# self.comboBox.cb.clear()
self.result_dic = []
def currentItem(self):
items = self.labelList.selectedItems()
if items:
return items[0]
return None
def currentBox(self):
items = self.BoxList.selectedItems()
if items:
return items[0]
return None
def addRecentFile(self, filePath):
if filePath in self.recentFiles:
self.recentFiles.remove(filePath)
elif len(self.recentFiles) >= self.maxRecent:
self.recentFiles.pop()
self.recentFiles.insert(0, filePath)
def beginner(self):
return self._beginner
def advanced(self):
return not self.beginner()
def getAvailableScreencastViewer(self):
osName = platform.system()
if osName == 'Windows':
return ['C:\\Program Files\\Internet Explorer\\iexplore.exe']
elif osName == 'Linux':
return ['xdg-open']
elif osName == 'Darwin':
return ['open']
## Callbacks ##
def showTutorialDialog(self):
subprocess.Popen(self.screencastViewer + [self.screencast])
def showInfoDialog(self):
from libs.__init__ import __version__
msg = u'Name:{0} \nApp Version:{1} \n{2} '.format(__appname__, __version__, sys.version_info)
QMessageBox.information(self, u'Information', msg)
def showStepsDialog(self):
msg = stepsInfo(self.lang)
QMessageBox.information(self, u'Information', msg)
def createShape(self):
assert self.beginner()
self.canvas.setEditing(False)
self.actions.create.setEnabled(False)
self.canvas.fourpoint = False
def createPolygon(self):
assert self.beginner()
self.canvas.setEditing(False)
self.canvas.fourpoint = True
self.actions.create.setEnabled(False)
def toggleDrawingSensitive(self, drawing=True):
"""In the middle of drawing, toggling between modes should be disabled."""
self.actions.editMode.setEnabled(not drawing)
if not drawing and self.beginner():
# Cancel creation.
print('Cancel creation.')
self.canvas.setEditing(True)
self.canvas.restoreCursor()
self.actions.create.setEnabled(True)
def toggleDrawMode(self, edit=True):
self.canvas.setEditing(edit)
self.actions.createMode.setEnabled(edit)
self.actions.editMode.setEnabled(not edit)
def setCreateMode(self):
assert self.advanced()
self.toggleDrawMode(False)
def setEditMode(self):
assert self.advanced()
self.toggleDrawMode(True)
self.labelSelectionChanged()
def updateFileMenu(self):
currFilePath = self.filePath
def exists(filename):
return os.path.exists(filename)
menu = self.menus.recentFiles
menu.clear()
files = [f for f in self.recentFiles if f !=
currFilePath and exists(f)]
for i, f in enumerate(files):
icon = newIcon('labels')
action = QAction(
icon, '&%d %s' % (i + 1, QFileInfo(f).fileName()), self)
action.triggered.connect(partial(self.loadRecent, f))
menu.addAction(action)
def popLabelListMenu(self, point):
self.menus.labelList.exec_(self.labelList.mapToGlobal(point))
def editLabel(self):
if not self.canvas.editing():
return
item = self.currentItem()
if not item:
return
text = self.labelDialog.popUp(item.text())
if text is not None:
item.setText(text)
# item.setBackground(generateColorByText(text))
self.setDirty()
self.updateComboBox()
######## detection box related functions #######
def boxItemChanged(self, item):
shape = self.itemsToShapesbox[item]
box = ast.literal_eval(item.text())
# print('shape in labelItemChanged is',shape.points)
if box != [(p.x(), p.y()) for p in shape.points]:
# shape.points = box
shape.points = [QPointF(p[0], p[1]) for p in box]
# QPointF(x,y)
# shape.line_color = generateColorByText(shape.label)
self.setDirty()
else: # User probably changed item visibility
self.canvas.setShapeVisible(shape, True)#item.checkState() == Qt.Checked
def editBox(self): # ADD
if not self.canvas.editing():
return
item = self.currentBox()
if not item:
return
text = self.labelDialog.popUp(item.text())
imageSize = str(self.image.size())
width, height = self.image.width(), self.image.height()
if text:
try:
text_list = eval(text)
except:
msg_box = QMessageBox(QMessageBox.Warning, 'Warning', 'Please enter the correct format')
msg_box.exec_()
return
if len(text_list) < 4:
msg_box = QMessageBox(QMessageBox.Warning, 'Warning', 'Please enter the coordinates of 4 points')
msg_box.exec_()
return
for box in text_list:
if box[0] > width or box[0] < 0 or box[1] > height or box[1] < 0:
msg_box = QMessageBox(QMessageBox.Warning, 'Warning', 'Out of picture size')
msg_box.exec_()
return
item.setText(text)
# item.setBackground(generateColorByText(text))
self.setDirty()
self.updateComboBox()
def updateBoxlist(self):
shape = self.canvas.selectedShape
item = self.shapesToItemsbox[shape] # listitem
text = [(int(p.x()), int(p.y())) for p in shape.points]
item.setText(str(text))
self.setDirty()
def indexTo5Files(self, currIndex):
if currIndex < 2:
return self.mImgList[:5]
elif currIndex > len(self.mImgList)-3:
return self.mImgList[-5:]
else:
return self.mImgList[currIndex - 2 : currIndex + 3]
# Tzutalin 20160906 : Add file list and dock to move faster
def fileitemDoubleClicked(self, item=None):
self.currIndex = self.mImgList.index(ustr(os.path.join(os.path.abspath(self.dirname), item.text())))
filename = self.mImgList[self.currIndex]
if filename:
self.mImgList5 = self.indexTo5Files(self.currIndex)
# self.additems5(None)
self.loadFile(filename)
def iconitemDoubleClicked(self, item=None):
self.currIndex = self.mImgList.index(ustr(os.path.join(item.toolTip())))
filename = self.mImgList[self.currIndex]
if filename:
self.mImgList5 = self.indexTo5Files(self.currIndex)
# self.additems5(None)
self.loadFile(filename)
def CanvasSizeChange(self):
if len(self.mImgList) > 0:
self.zoomWidget.setValue(self.zoomWidgetValue + self.imgsplider.value())
# React to canvas signals.
def shapeSelectionChanged(self, selected=False):
if self._noSelectionSlot:
self._noSelectionSlot = False
else:
shape = self.canvas.selectedShape
if shape:
self.shapesToItems[shape].setSelected(True)
self.shapesToItemsbox[shape].setSelected(True) # ADD
else:
self.labelList.clearSelection()
self.actions.delete.setEnabled(selected)
self.actions.copy.setEnabled(selected)
self.actions.edit.setEnabled(selected)
self.actions.shapeLineColor.setEnabled(selected)
self.actions.shapeFillColor.setEnabled(selected)
def addLabel(self, shape):
shape.paintLabel = self.displayLabelOption.isChecked()
item = HashableQListWidgetItem(shape.label)
item.setFlags(item.flags() | Qt.ItemIsUserCheckable)
item.setCheckState(Qt.Unchecked) if shape.difficult else item.setCheckState(Qt.Checked)
# Checked means difficult is False
# item.setBackground(generateColorByText(shape.label))
self.itemsToShapes[item] = shape
self.shapesToItems[shape] = item
self.labelList.addItem(item)
# print('item in add label is ',[(p.x(), p.y()) for p in shape.points], shape.label)
# ADD for box
item = HashableQListWidgetItem(str([(int(p.x()), int(p.y())) for p in shape.points]))
self.itemsToShapesbox[item] = shape
self.shapesToItemsbox[shape] = item
self.BoxList.addItem(item)
for action in self.actions.onShapesPresent:
action.setEnabled(True)
self.updateComboBox()
def remLabel(self, shape):
if shape is None:
# print('rm empty label')
return
item = self.shapesToItems[shape]
self.labelList.takeItem(self.labelList.row(item))
del self.shapesToItems[shape]
del self.itemsToShapes[item]
self.updateComboBox()
# ADD:
item = self.shapesToItemsbox[shape]
self.BoxList.takeItem(self.BoxList.row(item))
del self.shapesToItemsbox[shape]
del self.itemsToShapesbox[item]
self.updateComboBox()
def loadLabels(self, shapes):
s = []
for label, points, line_color, fill_color, difficult in shapes:
shape = Shape(label=label)
for x, y in points:
# Ensure the labels are within the bounds of the image. If not, fix them.
x, y, snapped = self.canvas.snapPointToCanvas(x, y)
if snapped:
self.setDirty()
shape.addPoint(QPointF(x, y))
shape.difficult = difficult
shape.close()
s.append(shape)
# if line_color:
# shape.line_color = QColor(*line_color)
# else:
# shape.line_color = generateColorByText(label)
#
# if fill_color:
# shape.fill_color = QColor(*fill_color)
# else:
# shape.fill_color = generateColorByText(label)
self.addLabel(shape)
self.updateComboBox()
self.canvas.loadShapes(s)
def updateComboBox(self):
# Get the unique labels and add them to the Combobox.
itemsTextList = [str(self.labelList.item(i).text()) for i in range(self.labelList.count())]
uniqueTextList = list(set(itemsTextList))
# Add a null row for showing all the labels
uniqueTextList.append("")
uniqueTextList.sort()
# self.comboBox.update_items(uniqueTextList)
def saveLabels(self, annotationFilePath, mode='Auto'):
# Mode is Auto means that labels will be loaded from self.result_dic totally, which is the output of ocr model
annotationFilePath = ustr(annotationFilePath)
if self.labelFile is None:
self.labelFile = LabelFile()
self.labelFile.verified = self.canvas.verified
def format_shape(s):
# print('s in saveLabels is ',s)
return dict(label=s.label, # str
line_color=s.line_color.getRgb(),
fill_color=s.fill_color.getRgb(),
points=[(p.x(), p.y()) for p in s.points], # QPonitF
# add chris
difficult=s.difficult) # bool
shapes = [] if mode == 'Auto' else \
[format_shape(shape) for shape in self.canvas.shapes]
# Can add differrent annotation formats here
for box in self.result_dic:
trans_dic = {"label": box[1][0], "points": box[0], 'difficult': False}
if trans_dic["label"] is "" and mode == 'Auto':
continue
shapes.append(trans_dic)
try:
trans_dic = []
for box in shapes:
trans_dic.append({"transcription": box['label'], "points": box['points'], 'difficult': box['difficult']})
self.PPlabel[annotationFilePath] = trans_dic
if mode == 'Auto':
self.Cachelabel[annotationFilePath] = trans_dic
# else:
# self.labelFile.save(annotationFilePath, shapes, self.filePath, self.imageData,
# self.lineColor.getRgb(), self.fillColor.getRgb())
# print('Image:{0} -> Annotation:{1}'.format(self.filePath, annotationFilePath))
return True
except LabelFileError as e:
self.errorMessage(u'Error saving label data', u'<b>%s</b>' % e)
return False
def copySelectedShape(self):
self.addLabel(self.canvas.copySelectedShape())
# fix copy and delete
self.shapeSelectionChanged(True)
def labelSelectionChanged(self):
item = self.currentItem()
if item and self.canvas.editing():
self._noSelectionSlot = True
self.canvas.selectShape(self.itemsToShapes[item])
shape = self.itemsToShapes[item]
def boxSelectionChanged(self):
item = self.currentBox()
if item and self.canvas.editing():
self._noSelectionSlot = True
self.canvas.selectShape(self.itemsToShapesbox[item])
shape = self.itemsToShapesbox[item]
def labelItemChanged(self, item):
shape = self.itemsToShapes[item]
label = item.text()
if label != shape.label:
shape.label = item.text()
# shape.line_color = generateColorByText(shape.label)
self.setDirty()
elif not ((item.checkState()== Qt.Unchecked) ^ (not shape.difficult)):
shape.difficult = True if item.checkState() == Qt.Unchecked else False
self.setDirty()
else: # User probably changed item visibility
self.canvas.setShapeVisible(shape, True) # item.checkState() == Qt.Checked
# self.actions.save.setEnabled(True)
# Callback functions:
def newShape(self):
"""Pop-up and give focus to the label editor.
position MUST be in global coordinates.
"""
if len(self.labelHist) > 0:
self.labelDialog = LabelDialog(
parent=self, listItem=self.labelHist)
# Sync single class mode from PR#106
if self.singleClassMode.isChecked() and self.lastLabel:
text = self.lastLabel
else:
text = self.labelDialog.popUp(text=self.prevLabelText)
self.lastLabel = text
if text is not None:
self.prevLabelText = self.stringBundle.getString('tempLabel')
# generate_color = generateColorByText(text)
shape = self.canvas.setLastLabel(text, None, None)#generate_color, generate_color
self.addLabel(shape)
if self.beginner(): # Switch to edit mode.
self.canvas.setEditing(True)
self.actions.create.setEnabled(True)
else:
self.actions.editMode.setEnabled(True)
self.setDirty()
else:
# self.canvas.undoLastLine()
self.canvas.resetAllLines()
def scrollRequest(self, delta, orientation):
units = - delta / (8 * 15)
bar = self.scrollBars[orientation]
bar.setValue(bar.value() + bar.singleStep() * units)
def setZoom(self, value):
self.actions.fitWidth.setChecked(False)
self.actions.fitWindow.setChecked(False)
self.zoomMode = self.MANUAL_ZOOM
self.zoomWidget.setValue(value)
def addZoom(self, increment=10):
self.setZoom(self.zoomWidget.value() + increment)
def zoomRequest(self, delta):
# get the current scrollbar positions
# calculate the percentages ~ coordinates
h_bar = self.scrollBars[Qt.Horizontal]
v_bar = self.scrollBars[Qt.Vertical]
# get the current maximum, to know the difference after zooming
h_bar_max = h_bar.maximum()
v_bar_max = v_bar.maximum()
# get the cursor position and canvas size
# calculate the desired movement from 0 to 1
# where 0 = move left
# 1 = move right
# up and down analogous
cursor = QCursor()
pos = cursor.pos()
relative_pos = QWidget.mapFromGlobal(self, pos)
cursor_x = relative_pos.x()
cursor_y = relative_pos.y()
w = self.scrollArea.width()
h = self.scrollArea.height()
# the scaling from 0 to 1 has some padding
# you don't have to hit the very leftmost pixel for a maximum-left movement
margin = 0.1
move_x = (cursor_x - margin * w) / (w - 2 * margin * w)
move_y = (cursor_y - margin * h) / (h - 2 * margin * h)
# clamp the values from 0 to 1
move_x = min(max(move_x, 0), 1)
move_y = min(max(move_y, 0), 1)
# zoom in
units = delta / (8 * 15)
scale = 10
self.addZoom(scale * units)
# get the difference in scrollbar values
# this is how far we can move
d_h_bar_max = h_bar.maximum() - h_bar_max
d_v_bar_max = v_bar.maximum() - v_bar_max
# get the new scrollbar values
new_h_bar_value = h_bar.value() + move_x * d_h_bar_max
new_v_bar_value = v_bar.value() + move_y * d_v_bar_max
h_bar.setValue(new_h_bar_value)
v_bar.setValue(new_v_bar_value)
def setFitWindow(self, value=True):
if value:
self.actions.fitWidth.setChecked(False)
self.zoomMode = self.FIT_WINDOW if value else self.MANUAL_ZOOM
self.adjustScale()
def setFitWidth(self, value=True):
if value:
self.actions.fitWindow.setChecked(False)
self.zoomMode = self.FIT_WIDTH if value else self.MANUAL_ZOOM
self.adjustScale()
def togglePolygons(self, value):
for item, shape in self.itemsToShapes.items():
self.canvas.setShapeVisible(shape, value)
def loadFile(self, filePath=None):
"""Load the specified file, or the last opened file if None."""
self.resetState()
self.canvas.setEnabled(False)
if filePath is None:
filePath = self.settings.get(SETTING_FILENAME)
# Make sure that filePath is a regular python string, rather than QString
filePath = ustr(filePath)
# Fix bug: An index error after select a directory when open a new file.
unicodeFilePath = ustr(filePath)
# unicodeFilePath = os.path.abspath(unicodeFilePath)
# Tzutalin 20160906 : Add file list and dock to move faster
# Highlight the file item
if unicodeFilePath and self.fileListWidget.count() > 0:
if unicodeFilePath in self.mImgList:
index = self.mImgList.index(unicodeFilePath)
fileWidgetItem = self.fileListWidget.item(index)
print('unicodeFilePath is', unicodeFilePath)
fileWidgetItem.setSelected(True)
###
self.iconlist.clear()
self.additems5(None)
for i in range(5):
item_tooltip = self.iconlist.item(i).toolTip()
# print(i,"---",item_tooltip)
if item_tooltip == ustr(filePath):
titem = self.iconlist.item(i)
titem.setSelected(True)
self.iconlist.scrollToItem(titem)
break
else:
self.fileListWidget.clear()
self.mImgList.clear()
self.iconlist.clear()
# if unicodeFilePath and self.iconList.count() > 0:
# if unicodeFilePath in self.mImgList:
if unicodeFilePath and os.path.exists(unicodeFilePath):
if LabelFile.isLabelFile(unicodeFilePath):
try:
self.labelFile = LabelFile(unicodeFilePath)
except LabelFileError as e:
self.errorMessage(u'Error opening file',
(u"<p><b>%s</b></p>"
u"<p>Make sure <i>%s</i> is a valid label file.")
% (e, unicodeFilePath))
self.status("Error reading %s" % unicodeFilePath)
return False
self.imageData = self.labelFile.imageData
self.lineColor = QColor(*self.labelFile.lineColor)
self.fillColor = QColor(*self.labelFile.fillColor)
self.canvas.verified = self.labelFile.verified
else:
# Load image:
# read data first and store for saving into label file.
self.imageData = read(unicodeFilePath, None)
self.labelFile = None
self.canvas.verified = False
image = QImage.fromData(self.imageData)
if image.isNull():
self.errorMessage(u'Error opening file',
u"<p>Make sure <i>%s</i> is a valid image file." % unicodeFilePath)
self.status("Error reading %s" % unicodeFilePath)
return False
self.status("Loaded %s" % os.path.basename(unicodeFilePath))
self.image = image
self.filePath = unicodeFilePath
self.canvas.loadPixmap(QPixmap.fromImage(image))
if self.labelFile:
self.loadLabels(self.labelFile.shapes)
if self.validFilestate(filePath) is True:
self.setClean()
else:
self.dirty = False
self.actions.save.setEnabled(True)
self.canvas.setEnabled(True)
self.adjustScale(initial=True)
self.paintCanvas()
self.addRecentFile(self.filePath)
self.toggleActions(True)
self.showBoundingBoxFromPPlabel(filePath)
self.setWindowTitle(__appname__ + ' ' + filePath)
# Default : select last item if there is at least one item
if self.labelList.count():
self.labelList.setCurrentItem(self.labelList.item(self.labelList.count() - 1))
self.labelList.item(self.labelList.count() - 1).setSelected(True)
self.canvas.setFocus(True)
return True
return False
def showBoundingBoxFromPPlabel(self, filePath):
imgidx = self.getImglabelidx(filePath)
if imgidx not in self.PPlabel.keys():
return
shapes = []
for box in self.PPlabel[imgidx]:
shapes.append((box['transcription'], box['points'], None, None, box['difficult']))
print(shapes)
self.loadLabels(shapes)
self.canvas.verified = False
def validFilestate(self, filePath):
if filePath not in self.fileStatedict.keys():
return None
elif self.fileStatedict[filePath] == 1:
return True
else:
return False
def resizeEvent(self, event):
if self.canvas and not self.image.isNull() \
and self.zoomMode != self.MANUAL_ZOOM:
self.adjustScale()
super(MainWindow, self).resizeEvent(event)
def paintCanvas(self):
assert not self.image.isNull(), "cannot paint null image"
self.canvas.scale = 0.01 * self.zoomWidget.value()
self.canvas.adjustSize()
self.canvas.update()
def adjustScale(self, initial=False):
value = self.scalers[self.FIT_WINDOW if initial else self.zoomMode]()
self.zoomWidget.setValue(int(100 * value))
def scaleFitWindow(self):
"""Figure out the size of the pixmap in order to fit the main widget."""
e = 2.0 # So that no scrollbars are generated.
w1 = self.centralWidget().width() - e
h1 = self.centralWidget().height() - e -110
a1 = w1 / h1
# Calculate a new scale value based on the pixmap's aspect ratio.
w2 = self.canvas.pixmap.width() - 0.0
h2 = self.canvas.pixmap.height() - 0.0
a2 = w2 / h2
return w1 / w2 if a2 >= a1 else h1 / h2
def scaleFitWidth(self):
# The epsilon does not seem to work too well here.
w = self.centralWidget().width() - 2.0
return w / self.canvas.pixmap.width()
def closeEvent(self, event):
if not self.mayContinue():
event.ignore()
else:
settings = self.settings
# If it loads images from dir, don't load it at the begining
if self.dirname is None:
settings[SETTING_FILENAME] = self.filePath if self.filePath else ''
else:
settings[SETTING_FILENAME] = ''
settings[SETTING_WIN_SIZE] = self.size()
settings[SETTING_WIN_POSE] = self.pos()
settings[SETTING_WIN_STATE] = self.saveState()
settings[SETTING_LINE_COLOR] = self.lineColor
settings[SETTING_FILL_COLOR] = self.fillColor
settings[SETTING_RECENT_FILES] = self.recentFiles
settings[SETTING_ADVANCE_MODE] = not self._beginner
if self.defaultSaveDir and os.path.exists(self.defaultSaveDir):
settings[SETTING_SAVE_DIR] = ustr(self.defaultSaveDir)
else:
settings[SETTING_SAVE_DIR] = ''
if self.lastOpenDir and os.path.exists(self.lastOpenDir):
settings[SETTING_LAST_OPEN_DIR] = self.lastOpenDir
else:
settings[SETTING_LAST_OPEN_DIR] = ''
settings[SETTING_SINGLE_CLASS] = self.singleClassMode.isChecked()
settings[SETTING_PAINT_LABEL] = self.displayLabelOption.isChecked()
settings[SETTING_DRAW_SQUARE] = self.drawSquaresOption.isChecked()
settings.save()
try:
self.saveFilestate()
self.savePPlabel()
except:
pass
def loadRecent(self, filename):
if self.mayContinue():
self.loadFile(filename)
def scanAllImages(self, folderPath):
extensions = ['.%s' % fmt.data().decode("ascii").lower() for fmt in QImageReader.supportedImageFormats()]
images = []
for file in os.listdir(folderPath):
if file.lower().endswith(tuple(extensions)):
relativePath = os.path.join(folderPath, file)
path = ustr(os.path.abspath(relativePath))
images.append(path)
natural_sort(images, key=lambda x: x.lower())
return images
def openDirDialog(self, _value=False, dirpath=None, silent=False):
if not self.mayContinue():
return
defaultOpenDirPath = dirpath if dirpath else '.'
if self.lastOpenDir and os.path.exists(self.lastOpenDir):
defaultOpenDirPath = self.lastOpenDir
else:
defaultOpenDirPath = os.path.dirname(self.filePath) if self.filePath else '.'
if silent != True:
targetDirPath = ustr(QFileDialog.getExistingDirectory(self,
'%s - Open Directory' % __appname__,
defaultOpenDirPath,
QFileDialog.ShowDirsOnly | QFileDialog.DontResolveSymlinks))
else:
targetDirPath = ustr(defaultOpenDirPath)
self.lastOpenDir = targetDirPath
self.importDirImages(targetDirPath)
def importDirImages(self, dirpath, isDelete = False):
if not self.mayContinue() or not dirpath:
return
if self.defaultSaveDir and self.defaultSaveDir != dirpath:
self.saveFilestate()
self.savePPlabel()
if not isDelete:
self.loadFilestate(dirpath)
self.PPlabelpath = dirpath+ '/Label.txt'
self.PPlabel = self.loadLabelFile(self.PPlabelpath)
self.Cachelabelpath = dirpath + '/Cache.cach'
self.Cachelabel = self.loadLabelFile(self.Cachelabelpath)
if self.Cachelabel:
self.PPlabel = dict(self.Cachelabel, **self.PPlabel)
self.lastOpenDir = dirpath
self.dirname = dirpath
self.defaultSaveDir = dirpath
self.statusBar().showMessage('%s started. Annotation will be saved to %s' %
(__appname__, self.defaultSaveDir))
self.statusBar().show()
self.filePath = None
self.fileListWidget.clear()
self.mImgList = self.scanAllImages(dirpath)
self.mImgList5 = self.mImgList[:5]
self.openNextImg()
doneicon = newIcon('done')
closeicon = newIcon('close')
for imgPath in self.mImgList:
filename = os.path.basename(imgPath)
if self.validFilestate(imgPath) is True:
item = QListWidgetItem(doneicon, filename)
else:
item = QListWidgetItem(closeicon, filename)
self.fileListWidget.addItem(item)
print('dirPath in importDirImages is', dirpath)
self.iconlist.clear()
self.additems5(dirpath)
self.changeFileFolder = True
self.haveAutoReced = False
self.AutoRecognition.setEnabled(True)
self.reRecogButton.setEnabled(True)
def verifyImg(self, _value=False):
# Proceding next image without dialog if having any label
if self.filePath is not None:
try:
self.labelFile.toggleVerify()
except AttributeError:
# If the labelling file does not exist yet, create if and
# re-save it with the verified attribute.
self.saveFile()
if self.labelFile != None:
self.labelFile.toggleVerify()
else:
return
self.canvas.verified = self.labelFile.verified
self.paintCanvas()
self.saveFile()
def openPrevImg(self, _value=False):
if len(self.mImgList) <= 0:
return
if self.filePath is None:
return
currIndex = self.mImgList.index(self.filePath)
self.mImgList5 = self.mImgList[:5]
if currIndex - 1 >= 0:
filename = self.mImgList[currIndex - 1]
self.mImgList5 = self.indexTo5Files(currIndex - 1)
if filename:
self.loadFile(filename)
def openNextImg(self, _value=False):
if not self.mayContinue():
return
if len(self.mImgList) <= 0:
return
filename = None
if self.filePath is None:
filename = self.mImgList[0]
self.mImgList5 = self.mImgList[:5]
else:
currIndex = self.mImgList.index(self.filePath)
if currIndex + 1 < len(self.mImgList):
filename = self.mImgList[currIndex + 1]
self.mImgList5 = self.indexTo5Files(currIndex + 1)
else:
self.mImgList5 = self.indexTo5Files(currIndex)
if filename:
print('file name in openNext is ',filename)
self.loadFile(filename)
def openFile(self, _value=False):
if not self.mayContinue():
return
path = os.path.dirname(ustr(self.filePath)) if self.filePath else '.'
formats = ['*.%s' % fmt.data().decode("ascii").lower() for fmt in QImageReader.supportedImageFormats()]
filters = "Image & Label files (%s)" % ' '.join(formats + ['*%s' % LabelFile.suffix])
filename = QFileDialog.getOpenFileName(self, '%s - Choose Image or Label file' % __appname__, path, filters)
if filename:
if isinstance(filename, (tuple, list)):
filename = filename[0]
self.loadFile(filename)
# print('filename in openfile is ', self.filePath)
self.filePath = None
self.fileListWidget.clear()
self.iconlist.clear()
self.mImgList = [filename]
self.openNextImg()
if self.validFilestate(filename) is True:
item = QListWidgetItem(newIcon('done'), filename)
self.setClean()
elif self.validFilestate(filename) is None:
item = QListWidgetItem(newIcon('close'), filename)
else:
item = QListWidgetItem(newIcon('close'), filename)
self.setDirty()
self.fileListWidget.addItem(filename)
self.additems5(None)
print('opened image is', filename)
def updateFileListIcon(self, filename):
pass
def saveFile(self, _value=False, mode='Manual'):
# Manual mode is used for users click "Save" manually,which will change the state of the image
if self.defaultSaveDir is not None and len(ustr(self.defaultSaveDir)):
if self.filePath:
imgidx = self.getImglabelidx(self.filePath)
self._saveFile(imgidx, mode=mode)
else:
imgFileDir = os.path.dirname(self.filePath)
imgFileName = os.path.basename(self.filePath)
savedFileName = os.path.splitext(imgFileName)[0]
savedPath = os.path.join(imgFileDir, savedFileName)
self._saveFile(savedPath if self.labelFile
else self.saveFileDialog(removeExt=False), mode=mode)
def saveFileAs(self, _value=False):
assert not self.image.isNull(), "cannot save empty image"
self._saveFile(self.saveFileDialog())
def saveFileDialog(self, removeExt=True):
caption = '%s - Choose File' % __appname__
filters = 'File (*%s)' % LabelFile.suffix
openDialogPath = self.currentPath()
dlg = QFileDialog(self, caption, openDialogPath, filters)
dlg.setDefaultSuffix(LabelFile.suffix[1:])
dlg.setAcceptMode(QFileDialog.AcceptSave)
filenameWithoutExtension = os.path.splitext(self.filePath)[0]
dlg.selectFile(filenameWithoutExtension)
dlg.setOption(QFileDialog.DontUseNativeDialog, False)
if dlg.exec_():
fullFilePath = ustr(dlg.selectedFiles()[0])
if removeExt:
return os.path.splitext(fullFilePath)[0] # Return file path without the extension.
else:
return fullFilePath
return ''
def _saveFile(self, annotationFilePath, mode='Manual'):
if mode == 'Manual':
if annotationFilePath and self.saveLabels(annotationFilePath, mode=mode):
self.setClean()
self.statusBar().showMessage('Saved to %s' % annotationFilePath)
self.statusBar().show()
currIndex = self.mImgList.index(self.filePath)
item = self.fileListWidget.item(currIndex)
item.setIcon(newIcon('done'))
self.fileStatedict[self.filePath] = 1
if len(self.fileStatedict)%self.autoSaveNum ==0:
self.saveFilestate()
self.savePPlabel(mode='Auto')
self.fileListWidget.insertItem(int(currIndex), item)
self.openNextImg()
self.actions.saveRec.setEnabled(True)
elif mode == 'Auto':
if annotationFilePath and self.saveLabels(annotationFilePath, mode=mode):
self.setClean()
self.statusBar().showMessage('Saved to %s' % annotationFilePath)
self.statusBar().show()
def closeFile(self, _value=False):
if not self.mayContinue():
return
self.resetState()
self.setClean()
self.toggleActions(False)
self.canvas.setEnabled(False)
self.actions.saveAs.setEnabled(False)
def deleteImg(self):
deletePath = self.filePath
if deletePath is not None:
deleteInfo = self.deleteImgDialog()
if deleteInfo == QMessageBox.Yes:
if platform.system() == 'Windows':
from win32com.shell import shell, shellcon
shell.SHFileOperation((0, shellcon.FO_DELETE, deletePath, None,
shellcon.FOF_SILENT | shellcon.FOF_ALLOWUNDO | shellcon.FOF_NOCONFIRMATION,
None, None))
# linux
elif platform.system() == 'Linux':
cmd = 'trash ' + deletePath
os.system(cmd)
# macOS
elif platform.system() == 'Darwin':
import subprocess
absPath = os.path.abspath(deletePath).replace('\\', '\\\\').replace('"', '\\"')
cmd = ['osascript', '-e',
'tell app "Finder" to move {the POSIX file "' + absPath + '"} to trash']
print(cmd)
subprocess.call(cmd, stdout=open(os.devnull, 'w'))
if self.filePath in self.fileStatedict.keys():
self.fileStatedict.pop(self.filePath)
imgidx = self.getImglabelidx(self.filePath)
if imgidx in self.PPlabel.keys():
self.PPlabel.pop(imgidx)
self.openNextImg()
self.importDirImages(self.lastOpenDir, isDelete=True)
def deleteImgDialog(self):
yes, cancel = QMessageBox.Yes, QMessageBox.Cancel
msg = u'The image will be deleted to the recycle bin'
return QMessageBox.warning(self, u'Attention', msg, yes | cancel)
def resetAll(self):
self.settings.reset()
self.close()
proc = QProcess()
proc.startDetached(os.path.abspath(__file__))
def mayContinue(self): #
if not self.dirty:
return True
else:
discardChanges = self.discardChangesDialog()
if discardChanges == QMessageBox.No:
return True
elif discardChanges == QMessageBox.Yes:
self.saveFile()
return True
else:
return False
def discardChangesDialog(self):
yes, no, cancel = QMessageBox.Yes, QMessageBox.No, QMessageBox.Cancel
msg = u'You have unsaved changes, would you like to save them and proceed?\nClick "No" to undo all changes.'
return QMessageBox.warning(self, u'Attention', msg, yes | no | cancel)
def errorMessage(self, title, message):
return QMessageBox.critical(self, title,
'<p><b>%s</b></p>%s' % (title, message))
def currentPath(self):
return os.path.dirname(self.filePath) if self.filePath else '.'
def chooseColor1(self):
color = self.colorDialog.getColor(self.lineColor, u'Choose line color',
default=DEFAULT_LINE_COLOR)
if color:
self.lineColor = color
Shape.line_color = color
self.canvas.setDrawingColor(color)
self.canvas.update()
self.setDirty()
def deleteSelectedShape(self):
self.remLabel(self.canvas.deleteSelected())
self.setDirty()
if self.noShapes():
for action in self.actions.onShapesPresent:
action.setEnabled(False)
def chshapeLineColor(self):
color = self.colorDialog.getColor(self.lineColor, u'Choose line color',
default=DEFAULT_LINE_COLOR)
if color:
self.canvas.selectedShape.line_color = color
self.canvas.update()
self.setDirty()
def chshapeFillColor(self):
color = self.colorDialog.getColor(self.fillColor, u'Choose fill color',
default=DEFAULT_FILL_COLOR)
if color:
self.canvas.selectedShape.fill_color = color
self.canvas.update()
self.setDirty()
def copyShape(self):
self.canvas.endMove(copy=True)
self.addLabel(self.canvas.selectedShape)
self.setDirty()
def moveShape(self):
self.canvas.endMove(copy=False)
self.setDirty()
def loadPredefinedClasses(self, predefClassesFile):
if os.path.exists(predefClassesFile) is True:
with codecs.open(predefClassesFile, 'r', 'utf8') as f:
for line in f:
line = line.strip()
if self.labelHist is None:
self.labelHist = [line]
else:
self.labelHist.append(line)
def togglePaintLabelsOption(self):
for shape in self.canvas.shapes:
shape.paintLabel = self.displayLabelOption.isChecked()
def toogleDrawSquare(self):
self.canvas.setDrawingShapeToSquare(self.drawSquaresOption.isChecked())
def additems(self, dirpath):
for file in self.mImgList:
pix = QPixmap(file)
_, filename = os.path.split(file)
filename, _ = os.path.splitext(filename)
item = QListWidgetItem(QIcon(pix.scaled(100, 100, Qt.IgnoreAspectRatio, Qt.FastTransformation)),
filename[:10])
item.setToolTip(file)
self.iconlist.addItem(item)
def additems5(self, dirpath):
for file in self.mImgList5:
pix = QPixmap(file)
_, filename = os.path.split(file)
filename, _ = os.path.splitext(filename)
pfilename = filename[:10]
if len(pfilename) < 10:
lentoken = 12 - len(pfilename)
prelen = lentoken // 2
bfilename = prelen * " " + pfilename + (lentoken - prelen) * " "
# item = QListWidgetItem(QIcon(pix.scaled(100, 100, Qt.KeepAspectRatio, Qt.SmoothTransformation)),filename[:10])
item = QListWidgetItem(QIcon(pix.scaled(100, 100, Qt.IgnoreAspectRatio, Qt.FastTransformation)),pfilename)
# item.setForeground(QBrush(Qt.white))
item.setToolTip(file)
self.iconlist.addItem(item)
owidth = 0
for index in range(len(self.mImgList5)):
item = self.iconlist.item(index)
itemwidget = self.iconlist.visualItemRect(item)
owidth += itemwidget.width()
self.iconlist.setMinimumWidth(owidth + 50)
def getImglabelidx(self, filePath):
if platform.system()=='Windows':
spliter = '\\'
else:
spliter = '/'
filepathsplit = filePath.split(spliter)[-2:]
return filepathsplit[0] + '/' + filepathsplit[1]
def autoRecognition(self):
assert self.mImgList is not None
print('Using model from ', self.model)
uncheckedList = [i for i in self.mImgList if i not in self.fileStatedict.keys()]
self.autoDialog = AutoDialog(parent=self, ocr=self.ocr, mImgList=uncheckedList, lenbar=len(uncheckedList))
self.autoDialog.popUp()
self.currIndex=len(self.mImgList)
self.loadFile(self.filePath) # ADD
self.haveAutoReced = True
self.AutoRecognition.setEnabled(False)
self.setDirty()
self.saveCacheLabel()
def reRecognition(self):
img = cv2.imread(self.filePath)
# org_box = [dic['points'] for dic in self.PPlabel[self.getImglabelidx(self.filePath)]]
if self.canvas.shapes:
self.result_dic = []
rec_flag = 0
for shape in self.canvas.shapes:
box = [[int(p.x()), int(p.y())] for p in shape.points]
assert len(box) == 4
img_crop = get_rotate_crop_image(img, np.array(box, np.float32))
if img_crop is None:
msg = 'Can not recognise the detection box in ' + self.filePath + '. Please change manually'
QMessageBox.information(self, "Information", msg)
return
result = self.ocr.ocr(img_crop, cls=True, det=False)
if result[0][0] is not '':
result.insert(0, box)
print('result in reRec is ', result)
self.result_dic.append(result)
if result[1][0] == shape.label:
print('label no change')
else:
rec_flag += 1
if len(self.result_dic) > 0 and rec_flag > 0:
self.saveFile(mode='Auto')
self.loadFile(self.filePath)
self.setDirty()
elif len(self.result_dic) == len(self.canvas.shapes) and rec_flag == 0:
QMessageBox.information(self, "Information", "The recognition result remains unchanged!")
else:
print('Can not recgonise in ', self.filePath)
else:
QMessageBox.information(self, "Information", "Draw a box!")
def autolcm(self):
vbox = QVBoxLayout()
hbox = QHBoxLayout()
self.panel = QLabel()
self.panel.setText(self.stringBundle.getString('choseModelLg'))
self.panel.setAlignment(Qt.AlignLeft)
self.comboBox = QComboBox()
self.comboBox.setObjectName("comboBox")
self.comboBox.addItems(['Chinese & English', 'English', 'French', 'German', 'Korean', 'Japanese'])
vbox.addWidget(self.panel)
vbox.addWidget(self.comboBox)
self.dialog = QDialog()
self.dialog.resize(300, 100)
self.okBtn = QPushButton(self.stringBundle.getString('ok'))
self.cancelBtn = QPushButton(self.stringBundle.getString('cancel'))
self.okBtn.clicked.connect(self.modelChoose)
self.cancelBtn.clicked.connect(self.cancel)
self.dialog.setWindowTitle(self.stringBundle.getString('choseModelLg'))
hbox.addWidget(self.okBtn)
hbox.addWidget(self.cancelBtn)
vbox.addWidget(self.panel)
vbox.addLayout(hbox)
self.dialog.setLayout(vbox)
self.dialog.setWindowModality(Qt.ApplicationModal)
self.dialog.exec_()
if self.filePath:
self.AutoRecognition.setEnabled(True)
def modelChoose(self):
print(self.comboBox.currentText())
lg_idx = {'Chinese & English': 'ch', 'English': 'en', 'French': 'french', 'German': 'german',
'Korean': 'korean', 'Japanese': 'japan'}
del self.ocr
self.ocr = PaddleOCR(use_pdserving=False, use_angle_cls=True, det=True, cls=True, use_gpu=False,
lang=lg_idx[self.comboBox.currentText()])
self.dialog.close()
def cancel(self):
self.dialog.close()
def loadFilestate(self, saveDir):
self.fileStatepath = saveDir + '/fileState.txt'
self.fileStatedict = {}
if not os.path.exists(self.fileStatepath):
f = open(self.fileStatepath, 'w', encoding='utf-8')
else:
with open(self.fileStatepath, 'r', encoding='utf-8') as f:
states = f.readlines()
for each in states:
file, state = each.split('\t')
self.fileStatedict[file] = 1
def saveFilestate(self):
with open(self.fileStatepath, 'w', encoding='utf-8') as f:
for key in self.fileStatedict:
f.write(key + '\t')
f.write(str(self.fileStatedict[key]) + '\n')
def loadLabelFile(self, labelpath):
labeldict = {}
if not os.path.exists(labelpath):
f = open(labelpath, 'w', encoding='utf-8')
else:
with open(labelpath, 'r', encoding='utf-8') as f:
data = f.readlines()
for each in data:
file, label = each.split('\t')
if label:
label = label.replace('false', 'False')
label = label.replace('true', 'True')
labeldict[file] = eval(label)
else:
labeldict[file] = []
return labeldict
def savePPlabel(self,mode='Manual'):
savedfile = [self.getImglabelidx(i) for i in self.fileStatedict.keys()]
with open(self.PPlabelpath, 'w', encoding='utf-8') as f:
for key in self.PPlabel:
if key in savedfile and self.PPlabel[key] != []:
f.write(key + '\t')
f.write(json.dumps(self.PPlabel[key], ensure_ascii=False) + '\n')
if mode=='Manual':
msg = 'Images that have been checked are saved in '+ self.PPlabelpath
QMessageBox.information(self, "Information", msg)
def saveCacheLabel(self):
with open(self.Cachelabelpath, 'w', encoding='utf-8') as f:
for key in self.Cachelabel:
f.write(key + '\t')
f.write(json.dumps(self.Cachelabel[key], ensure_ascii=False) + '\n')
def saveRecResult(self):
if None in [self.PPlabelpath, self.PPlabel, self.fileStatedict]:
QMessageBox.information(self, "Information", "Save file first")
return
rec_gt_dir = os.path.dirname(self.PPlabelpath) + '/rec_gt.txt'
crop_img_dir = os.path.dirname(self.PPlabelpath) + '/crop_img/'
if not os.path.exists(crop_img_dir):
os.mkdir(crop_img_dir)
with open(rec_gt_dir, 'w', encoding='utf-8') as f:
for key in self.fileStatedict:
idx = self.getImglabelidx(key)
for i, label in enumerate(self.PPlabel[idx]):
if label['difficult']: continue
img = cv2.imread(key)
img_crop = get_rotate_crop_image(img, np.array(label['points'], np.float32))
img_name = os.path.splitext(os.path.basename(idx))[0] + '_crop_'+str(i)+'.jpg'
cv2.imwrite(crop_img_dir+img_name, img_crop)
f.write('crop_img/'+ img_name + '\t')
f.write(label['transcription'] + '\n')
QMessageBox.information(self, "Information", "Cropped images has been saved in "+str(crop_img_dir))
def inverted(color):
return QColor(*[255 - v for v in color.getRgb()])
def read(filename, default=None):
try:
with open(filename, 'rb') as f:
return f.read()
except:
return default
def get_main_app(argv=[]):
"""
Standard boilerplate Qt application code.
Do everything but app.exec_() -- so that we can test the application in one thread
"""
app = QApplication(argv)
app.setApplicationName(__appname__)
app.setWindowIcon(newIcon("app"))
# Tzutalin 201705+: Accept extra agruments to change predefined class file
argparser = argparse.ArgumentParser()
argparser.add_argument("--lang", default='ch', nargs="?")
argparser.add_argument("--predefined_classes_file",
default=os.path.join(os.path.dirname(__file__), "data", "predefined_classes.txt"),
nargs="?")
args = argparser.parse_args(argv[1:])
# Usage : labelImg.py image predefClassFile saveDir
win = MainWindow(lang=args.lang,
defaultPrefdefClassFile=args.predefined_classes_file)
win.show()
return app, win
def main():
'''construct main app and run it'''
app, _win = get_main_app(sys.argv)
return app.exec_()
if __name__ == '__main__':
resource_file = './libs/resources.py'
if not os.path.exists(resource_file):
output = os.system('pyrcc5 -o libs/resources.py resources.qrc')
assert output is 0, "operate the cmd have some problems ,please check whether there is a in the lib " \
"directory resources.py "
import libs.resources
sys.exit(main())
# PPOCRLabel
PPOCRLabel是一款适用于OCR领域的半自动化图形标注工具,使用python3和pyqt5编写,支持矩形框标注和四点标注模式,导出格式可直接用于PPOCR检测和识别模型的训练。
<img src="./data/gif/steps.gif" width="100%"/>
## 安装
### 1. 安装PaddleOCR
参考[PaddleOCR安装文档](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/doc/doc_ch/installation.md)准备好PaddleOCR
### 2. 安装PPOCRLabel
#### Windows + Anaconda
下载安装[Anaconda](https://www.anaconda.com/download/#download) (Python 3+)
```
conda install pyqt=5
cd ./PPOCRLabel # 将目录切换到PPOCRLabel文件夹下
pyrcc5 -o libs/resources.py resources.qrc
python PPOCRLabel.py
```
#### Ubuntu Linux
```
pip3 install pyqt5
pip3 install trash-cli
cd ./PPOCRLabel # 将目录切换到PPOCRLabel文件夹下
python3 PPOCRLabel.py
```
#### macOS
```
pip3 install pyqt5
pip3 uninstall opencv-python # 由于mac版本的opencv与pyqt有冲突,需先手动卸载opencv
pip3 install opencv-contrib-python-headless # 安装headless版本的open-cv
cd ./PPOCRLabel # 将目录切换到PPOCRLabel文件夹下
python3 PPOCRLabel.py
```
## 使用
### 操作步骤
1. 安装与运行:使用上述命令安装与运行程序。
2. 打开文件夹:在菜单栏点击 “文件” - "打开目录" 选择待标记图片的文件夹<sup>[1]</sup>.
3. 自动标注:点击 ”自动标注“,使用PPOCR超轻量模型对图片文件名前图片状态<sup>[2]</sup>为 “X” 的图片进行自动标注。
4. 手动标注:点击 “矩形标注”(推荐直接在英文模式下点击键盘中的 “W”),用户可对当前图片中模型未检出的部分进行手动绘制标记框。点击键盘P,则使用四点标注模式(或点击“编辑” - “四点标注”),用户依次点击4个点后,双击左键表示标注完成。
5. 标记框绘制完成后,用户点击 “确认”,检测框会先被预分配一个 “待识别” 标签。
6. 重新识别:将图片中的所有检测画绘制/调整完成后,点击 “重新识别”,PPOCR模型会对当前图片中的**所有检测框**重新识别<sup>[3]</sup>
7. 内容更改:双击识别结果,对不准确的识别结果进行手动更改。
8. 确认结果:点击 “确认”,图片状态切换为 “√”,跳转至下一张。
9. 删除:点击 “删除图像”,图片将会被删除至回收站。
10. 保存标注结果:关闭应用程序或切换文件路径后,手动确认过的标签将会被存放在所打开图片文件夹下的*Label.txt*中。在菜单栏点击 “PaddleOCR” - "保存识别结果"后,会将此类图片的识别训练数据保存在*crop_img*文件夹下,识别标签保存在*rec_gt.txt*<sup>[4]</sup>
### 注意
[1] PPOCRLabel以文件夹为基本标记单位,打开待标记的图片文件夹后,不会在窗口栏中显示图片,而是在点击 "选择文件夹" 之后直接将文件夹下的图片导入到程序中。
[2] 图片状态表示本张图片用户是否手动保存过,未手动保存过即为 “X”,手动保存过为 “√”。点击 “自动标注”按钮后,PPOCRLabel不会对状态为 “√” 的图片重新标注。
[3] 点击“重新识别”后,模型会对图片中的识别结果进行覆盖。因此如果在此之前手动更改过识别结果,有可能在重新识别后产生变动。
[4] PPOCRLabel产生的文件均在标记图片的文件夹中,包括一下几种,请勿手动更改其中内容,否则会引起程序出现异常。
| 文件名 | 说明 |
| :-----------: | :----------------------------------------------------------: |
| Label.txt | 检测标签,可直接用于PPOCR检测模型训练。用户每保存10张检测结果后,程序会进行自动写入。当用户关闭应用程序或切换文件路径后同样会进行写入。 |
| fileState.txt | 图片状态标记文件,保存当前文件夹下已经被用户手动确认过的图片名称。 |
| Cache.cach | 缓存文件,保存模型自动识别的结果。 |
| rec_gt.txt | 识别标签。可直接用于PPOCR识别模型训练。需用户手动点击菜单栏“PaddleOCR” - "保存识别结果"后产生。 |
| crop_img | 识别数据。按照检测框切割后的图片。与rec_gt.txt同时产生。 |
## 说明
### 内置模型
- 默认模型:PPOCRLabel默认使用PaddleOCR中的中英文超轻量OCR模型,支持中英文与数字识别,多种语言检测。
- 模型语言切换:用户可通过菜单栏中 "PaddleOCR" - "选择模型" 切换内置模型语言,目前支持的语言包括法文、德文、韩文、日文。具体模型下载链接可参考[PaddleOCR模型列表](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/doc/doc_ch/models_list.md).
- 自定义模型:用户可根据[自定义模型代码使用](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/doc/doc_ch/whl.md#%E8%87%AA%E5%AE%9A%E4%B9%89%E6%A8%A1%E5%9E%8B),通过修改PPOCRLabel.py中针对[PaddleOCR类的实例化](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/PPOCRLabel/PPOCRLabel.py#L110)替换成自己训练的模型。
### 导出部分识别结果
针对部分难以识别的数据,通过在识别结果的复选框中**取消勾选**相应的标记,其识别结果不会被导出。
*注意:识别结果中的复选框状态仍需用户手动点击保存后才能保留*
### 错误提示
- 如果同时使用whl包安装了paddleocr,其优先级大于通过paddleocr.py调用PaddleOCR类,whl包未更新时会导致程序异常。
- PPOCRLabel**不支持对中文文件名**的图片进行自动标注。
- 如果您在打开软件过程中出现**objc[XXXXX]**开头的错误,证明您的opencv版本太高,建议安装4.2版本:
```
pip install opencv-python==4.2.0.32
```
### 参考资料
1.[Tzutalin. LabelImg. Git code (2015)](https://github.com/tzutalin/labelImg)
# PPOCRLabel
PPOCRLabel is a semi-automatic graphic annotation tool suitable for OCR field. It is written in python3 and pyqt5, supporting rectangular box annotation and four-point annotation modes. Annotations can be directly used for the training of PPOCR detection and recognition models.
<img src="./data/gif/steps.gif" width="100%"/>
## Installation
### 1. Install PaddleOCR
Refer to [PaddleOCR installation document](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/doc/doc_ch/installation.md) to prepare PaddleOCR
### 2. Install PPOCRLabel
#### Windows + Anaconda
Download and install [Anaconda](https://www.anaconda.com/download/#download) (Python 3+)
```
conda install pyqt=5
cd ./PPOCRLabel # Change the directory to the PPOCRLabel folder
pyrcc5 -o libs/resources.py resources.qrc
python PPOCRLabel.py --lang en
```
#### Ubuntu Linux
```
pip3 install pyqt5
pip3 install trash-cli
cd ./PPOCRLabel # Change the directory to the PPOCRLabel folder
python3 PPOCRLabel.py --lang en
```
#### macOS
```
pip3 install pyqt5
pip3 uninstall opencv-python # Uninstall opencv manually as it conflicts with pyqt
pip3 install opencv-contrib-python-headless # Install the headless version of opencv
cd ./PPOCRLabel # Change the directory to the PPOCRLabel folder
python3 PPOCRLabel.py --lang en
```
## Usage
### Steps
1. Build and launch using the instructions above.
2. Click 'Open Dir' in Menu/File to select the folder of the picture.<sup>[1]</sup>
3. Click 'Auto recognition', use PPOCR model to automatically annotate images which marked with 'X' <sup>[2]</sup>before the file name.
4. Create Box:
4.1 Click 'Create RectBox' or press 'W' in English keyboard mode to draw a new rectangle detection box. Click and release left mouse to select a region to annotate the text area.
4.2 Press 'P' to enter four-point labeling mode which enables you to create any four-point shape by clicking four points with the left mouse button in succession and DOUBLE CLICK the left mouse as the signal of labeling completion.
5. After the marking frame is drawn, the user clicks "OK", and the detection frame will be pre-assigned a "TEMPORARY" label.
6. Click 're-Recognition', model will rewrite ALL recognition results in ALL detection box<sup>[3]</sup>.
7. Double click the result in 'recognition result' list to manually change inaccurate recognition results.
8. Click "Save", the image status will switch to "√",then the program automatically jump to the next.
9. Click "Delete Image" and the image will be deleted to the recycle bin.
10. Labeling result: After closing the application or switching the file path, the manually saved label will be stored in *Label.txt* under the opened picture folder.
Click "PaddleOCR"-"Save Recognition Results" in the menu bar, the recognition training data of such pictures will be saved in the *crop_img* folder, and the recognition label will be saved in *rec_gt.txt*<sup>[4]</sup>.
### Note
[1] PPOCRLabel uses the opened folder as the project. After opening the image folder, the picture will not be displayed in the dialog. Instead, the pictures under the folder will be directly imported into the program after clicking "Open Dir".
[2] The image status indicates whether the user has saved the image manually. If it has not been saved manually it is "X", otherwise it is "√", PPOCRLabel will not relabel pictures with a status of "√".
[3] After clicking "Re-recognize", the model will overwrite ALL recognition results in the picture.
Therefore, if the recognition result has been manually changed before, it may change after re-recognition.
[4] The files produced by PPOCRLabel include the following, please do not manually change the contents, otherwise it will cause the program to be abnormal.
| File name | Description |
| :-----------: | :----------------------------------------------------------: |
| Label.txt | The detection label file can be directly used for PPOCR detection model training. After the user saves 10 label results, the file will be automatically saved. It will also be written when the user closes the application or changes the file folder. |
| fileState.txt | The picture status file save the image in the current folder that has been manually confirmed by the user. |
| Cache.cach | Cache files to save the results of model recognition. |
| rec_gt.txt | The recognition label file, which can be directly used for PPOCR identification model training, is generated after the user clicks on the menu bar "PaddleOCR"-"Save recognition result". |
| crop_img | The recognition data, generated at the same time with *rec_gt.txt* |
## Explanation
### Built-in Model
- Default model: PPOCRLabel uses the Chinese and English ultra-lightweight OCR model in PaddleOCR by default, supports Chinese, English and number recognition, and multiple language detection.
- Model language switching: Changing the built-in model language is supportable by clicking "PaddleOCR"-"Choose OCR Model" in the menu bar. Currently supported languages​include French, German, Korean, and Japanese.
For specific model download links, please refer to [PaddleOCR Model List](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/doc/doc_en/models_list_en.md#multilingual-recognition-modelupdating)
- Custom model: The model trained by users can be replaced by modifying PPOCRLabel.py in [PaddleOCR class instantiation](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/PPOCRLabel/PPOCRLabel.py#L110) referring [Custom Model Code](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/doc/doc_en/whl_en.md#use-custom-model)
### Export partial recognition results
For some data that are difficult to recognize, the recognition results will not be exported by **unchecking** the corresponding tags in the recognition results checkbox.
*Note: The status of the checkboxes in the recognition results still needs to be saved manually by clicking Save Button.*
### Error message
- If paddleocr is installed with whl, it has a higher priority than calling PaddleOCR class with paddleocr.py, which may cause an exception if whl package is not updated.
- If you get an error starting with **objc[XXXXX]** when opening the software, it proves that your opencv version is too high. It is recommended to install version 4.2:
```
pip install opencv-python==4.2.0.32
```
### Related
1.[Tzutalin. LabelImg. Git code (2015)](https://github.com/tzutalin/labelImg)
# Copyright (c) <2015-Present> Tzutalin
# Copyright (C) 2013 MIT, Computer Science and Artificial Intelligence Laboratory. Bryan Russell, Antonio Torralba,
# William T. Freeman. Permission is hereby granted, free of charge, to any person obtaining a copy of this software and
# associated documentation files (the "Software"), to deal in the Software without restriction, including without
# limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the
# Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
# The above copyright notice and this permission notice shall be included in all copies or substantial portions of
# the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT
# NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT
# SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF
# CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
# THE SOFTWARE.
import sys
try:
from PyQt5.QtWidgets import QWidget, QHBoxLayout, QComboBox
except ImportError:
# needed for py3+qt4
# Ref:
# http://pyqt.sourceforge.net/Docs/PyQt4/incompatible_apis.html
# http://stackoverflow.com/questions/21217399/pyqt4-qtcore-qvariant-object-instead-of-a-string
if sys.version_info.major >= 3:
import sip
sip.setapi('QVariant', 2)
from PyQt4.QtGui import QWidget, QHBoxLayout, QComboBox
class ComboBox(QWidget):
def __init__(self, parent=None, items=[]):
super(ComboBox, self).__init__(parent)
layout = QHBoxLayout()
self.cb = QComboBox()
self.items = items
self.cb.addItems(self.items)
self.cb.currentIndexChanged.connect(parent.comboSelectionChanged)
layout.addWidget(self.cb)
self.setLayout(layout)
def update_items(self, items):
self.items = items
self.cb.clear()
self.cb.addItems(self.items)
__version_info__ = ('1', '0', '0')
__version__ = '.'.join(__version_info__)
try:
from PyQt5.QtGui import *
from PyQt5.QtCore import *
from PyQt5.QtWidgets import *
except ImportError:
from PyQt4.QtGui import *
from PyQt4.QtCore import *
import json
from libs.utils import newIcon
BB = QDialogButtonBox
class Worker(QThread):
progressBarValue = pyqtSignal(int)
listValue = pyqtSignal(str)
endsignal = pyqtSignal(int, str)
handle = 0
def __init__(self, ocr, mImgList, mainThread, model):
super(Worker, self).__init__()
self.ocr = ocr
self.mImgList = mImgList
self.mainThread = mainThread
self.model = model
self.setStackSize(1024*1024)
def run(self):
try:
findex = 0
for Imgpath in self.mImgList:
if self.handle == 0:
self.listValue.emit(Imgpath)
if self.model == 'paddle':
self.result_dic = self.ocr.ocr(Imgpath, cls=True, det=True)
# 结果保存
if self.result_dic is None or len(self.result_dic) == 0:
print('Can not recognise file is : ', Imgpath)
pass
else:
for res in self.result_dic:
chars = res[1][0]
cond = res[1][1]
posi = res[0]
self.listValue.emit("文字:" + chars + " 置信度:" + str(cond) + " 坐标:" + json.dumps(posi))
self.mainThread.result_dic = self.result_dic
self.mainThread.filePath = Imgpath
# 保存
self.mainThread.saveFile(mode='Auto')
findex += 1
self.progressBarValue.emit(findex)
else:
break
self.endsignal.emit(0, "readAll")
self.exec()
except Exception as e:
print(e)
raise
class AutoDialog(QDialog):
def __init__(self, text="Enter object label", parent=None, ocr=None, mImgList=None, lenbar=0):
super(AutoDialog, self).__init__(parent)
self.setFixedWidth(1000)
self.parent = parent
self.ocr = ocr
self.mImgList = mImgList
self.pb = QProgressBar()
self.pb.setRange(0, lenbar)
self.pb.setValue(0)
layout = QVBoxLayout()
layout.addWidget(self.pb)
self.model = 'paddle'
self.listWidget = QListWidget(self)
layout.addWidget(self.listWidget)
self.buttonBox = bb = BB(BB.Ok | BB.Cancel, Qt.Horizontal, self)
bb.button(BB.Ok).setIcon(newIcon('done'))
bb.button(BB.Cancel).setIcon(newIcon('undo'))
bb.accepted.connect(self.validate)
bb.rejected.connect(self.reject)
layout.addWidget(bb)
bb.button(BB.Ok).setEnabled(False)
self.setLayout(layout)
self.setWindowTitle("自动标注中")
self.setWindowModality(Qt.ApplicationModal)
# self.setWindowFlags(Qt.WindowCloseButtonHint)
self.thread_1 = Worker(self.ocr, self.mImgList, self.parent, 'paddle')
self.thread_1.progressBarValue.connect(self.handleProgressBarSingal)
self.thread_1.listValue.connect(self.handleListWidgetSingal)
self.thread_1.endsignal.connect(self.handleEndsignalSignal)
def handleProgressBarSingal(self, i):
self.pb.setValue(i)
def handleListWidgetSingal(self, i):
self.listWidget.addItem(i)
titem = self.listWidget.item(self.listWidget.count() - 1)
self.listWidget.scrollToItem(titem)
def handleEndsignalSignal(self, i, str):
if i == 0 and str == "readAll":
self.buttonBox.button(BB.Ok).setEnabled(True)
self.buttonBox.button(BB.Cancel).setEnabled(False)
def reject(self):
print("reject")
self.thread_1.handle = -1
self.thread_1.quit()
# del self.thread_1
# if self.thread_1.isRunning():
# self.thread_1.terminate()
# self.thread_1.quit()
# super(AutoDialog,self).reject()
while not self.thread_1.isFinished():
pass
self.accept()
def validate(self):
self.accept()
def postProcess(self):
try:
self.edit.setText(self.edit.text().trimmed())
# print(self.edit.text())
except AttributeError:
# PyQt5: AttributeError: 'str' object has no attribute 'trimmed'
self.edit.setText(self.edit.text())
print(self.edit.text())
def popUp(self):
self.thread_1.start()
return 1 if self.exec_() else None
def closeEvent(self, event):
print("???")
# if self.thread_1.isRunning():
# self.thread_1.quit()
#
# # self._thread.terminate()
# # del self.thread_1
# super(AutoDialog, self).closeEvent(event)
self.reject()
# Copyright (c) <2015-Present> Tzutalin
# Copyright (C) 2013 MIT, Computer Science and Artificial Intelligence Laboratory. Bryan Russell, Antonio Torralba,
# William T. Freeman. Permission is hereby granted, free of charge, to any person obtaining a copy of this software and
# associated documentation files (the "Software"), to deal in the Software without restriction, including without
# limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the
# Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
# The above copyright notice and this permission notice shall be included in all copies or substantial portions of
# the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT
# NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT
# SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF
# CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
# THE SOFTWARE.
try:
from PyQt5.QtGui import *
from PyQt5.QtCore import *
from PyQt5.QtWidgets import *
except ImportError:
from PyQt4.QtGui import *
from PyQt4.QtCore import *
#from PyQt4.QtOpenGL import *
from libs.shape import Shape
from libs.utils import distance
CURSOR_DEFAULT = Qt.ArrowCursor
CURSOR_POINT = Qt.PointingHandCursor
CURSOR_DRAW = Qt.CrossCursor
CURSOR_MOVE = Qt.ClosedHandCursor
CURSOR_GRAB = Qt.OpenHandCursor
# class Canvas(QGLWidget):
class Canvas(QWidget):
zoomRequest = pyqtSignal(int)
scrollRequest = pyqtSignal(int, int)
newShape = pyqtSignal()
selectionChanged = pyqtSignal(bool)
shapeMoved = pyqtSignal()
drawingPolygon = pyqtSignal(bool)
CREATE, EDIT = list(range(2))
_fill_drawing = False # draw shadows
epsilon = 11.0
def __init__(self, *args, **kwargs):
super(Canvas, self).__init__(*args, **kwargs)
# Initialise local state.
self.mode = self.EDIT
self.shapes = []
self.current = None
self.selectedShape = None # save the selected shape here
self.selectedShapeCopy = None
self.drawingLineColor = QColor(0, 0, 255)
self.drawingRectColor = QColor(0, 0, 255)
self.line = Shape(line_color=self.drawingLineColor)
self.prevPoint = QPointF()
self.offsets = QPointF(), QPointF()
self.scale = 1.0
self.pixmap = QPixmap()
self.visible = {}
self._hideBackround = False
self.hideBackround = False
self.hShape = None
self.hVertex = None
self._painter = QPainter()
self._cursor = CURSOR_DEFAULT
# Menus:
self.menus = (QMenu(), QMenu())
# Set widget options.
self.setMouseTracking(True)
self.setFocusPolicy(Qt.WheelFocus)
self.verified = False
self.drawSquare = False
self.fourpoint = True # ADD
self.pointnum = 0
#initialisation for panning
self.pan_initial_pos = QPoint()
def setDrawingColor(self, qColor):
self.drawingLineColor = qColor
self.drawingRectColor = qColor
def enterEvent(self, ev):
self.overrideCursor(self._cursor)
def leaveEvent(self, ev):
self.restoreCursor()
def focusOutEvent(self, ev):
self.restoreCursor()
def isVisible(self, shape):
return self.visible.get(shape, True)
def drawing(self):
return self.mode == self.CREATE
def editing(self):
return self.mode == self.EDIT
def setEditing(self, value=True):
self.mode = self.EDIT if value else self.CREATE
if not value: # Create
self.unHighlight()
self.deSelectShape()
self.prevPoint = QPointF()
self.repaint()
def unHighlight(self):
if self.hShape:
self.hShape.highlightClear()
self.hVertex = self.hShape = None
def selectedVertex(self):
return self.hVertex is not None
def mouseMoveEvent(self, ev):
"""Update line with last point and current coordinates."""
pos = self.transformPos(ev.pos())
# Update coordinates in status bar if image is opened
window = self.parent().window()
if window.filePath is not None:
self.parent().window().labelCoordinates.setText(
'X: %d; Y: %d' % (pos.x(), pos.y()))
# Polygon drawing.
if self.drawing():
self.overrideCursor(CURSOR_DRAW) # ?
if self.current:
# Display annotation width and height while drawing
currentWidth = abs(self.current[0].x() - pos.x())
currentHeight = abs(self.current[0].y() - pos.y())
self.parent().window().labelCoordinates.setText(
'Width: %d, Height: %d / X: %d; Y: %d' % (currentWidth, currentHeight, pos.x(), pos.y()))
color = self.drawingLineColor
if self.outOfPixmap(pos):
# Don't allow the user to draw outside the pixmap.
# Clip the coordinates to 0 or max,
# if they are outside the range [0, max]
size = self.pixmap.size()
clipped_x = min(max(0, pos.x()), size.width())
clipped_y = min(max(0, pos.y()), size.height())
pos = QPointF(clipped_x, clipped_y)
elif len(self.current) > 1 and self.closeEnough(pos, self.current[0]) and not self.fourpoint:
# Attract line to starting point and colorise to alert the
# user:
pos = self.current[0]
color = self.current.line_color
self.overrideCursor(CURSOR_POINT)
self.current.highlightVertex(0, Shape.NEAR_VERTEX)
elif ( # ADD
len(self.current) > 1
and self.fourpoint
and self.closeEnough(pos, self.current[0])
):
# Attract line to starting point and
# colorise to alert the user.
pos = self.current[0]
self.overrideCursor(CURSOR_POINT)
self.current.highlightVertex(0, Shape.NEAR_VERTEX)
if self.drawSquare:
initPos = self.current[0]
minX = initPos.x()
minY = initPos.y()
min_size = min(abs(pos.x() - minX), abs(pos.y() - minY))
directionX = -1 if pos.x() - minX < 0 else 1
directionY = -1 if pos.y() - minY < 0 else 1
self.line[1] = QPointF(minX + directionX * min_size, minY + directionY * min_size)
elif self.fourpoint:
# self.line[self.pointnum] = pos # OLD
self.line[0] = self.current[-1]
self.line[1] = pos
else:
self.line[1] = pos # pos is the mouse's current position
self.line.line_color = color
self.prevPoint = QPointF() # ?
self.current.highlightClear()
else:
self.prevPoint = pos
self.repaint()
return
# Polygon copy moving.
if Qt.RightButton & ev.buttons():
if self.selectedShapeCopy and self.prevPoint:
self.overrideCursor(CURSOR_MOVE)
self.boundedMoveShape(self.selectedShapeCopy, pos)
self.repaint()
elif self.selectedShape:
self.selectedShapeCopy = self.selectedShape.copy()
self.repaint()
return
# Polygon/Vertex moving.
if Qt.LeftButton & ev.buttons():
if self.selectedVertex():
self.boundedMoveVertex(pos)
self.shapeMoved.emit()
self.repaint()
elif self.selectedShape and self.prevPoint:
self.overrideCursor(CURSOR_MOVE)
self.boundedMoveShape(self.selectedShape, pos)
self.shapeMoved.emit()
self.repaint()
else:
#pan
delta_x = pos.x() - self.pan_initial_pos.x()
delta_y = pos.y() - self.pan_initial_pos.y()
self.scrollRequest.emit(delta_x, Qt.Horizontal)
self.scrollRequest.emit(delta_y, Qt.Vertical)
self.update()
return
# Just hovering over the canvas, 2 posibilities:
# - Highlight shapes
# - Highlight vertex
# Update shape/vertex fill and tooltip value accordingly.
self.setToolTip("Image")
for shape in reversed([s for s in self.shapes if self.isVisible(s)]):
# Look for a nearby vertex to highlight. If that fails,
# check if we happen to be inside a shape.
index = shape.nearestVertex(pos, self.epsilon)
if index is not None:
if self.selectedVertex():
self.hShape.highlightClear()
self.hVertex, self.hShape = index, shape
shape.highlightVertex(index, shape.MOVE_VERTEX)
self.overrideCursor(CURSOR_POINT)
self.setToolTip("Click & drag to move point")
self.setStatusTip(self.toolTip())
self.update()
break
elif shape.containsPoint(pos):
if self.selectedVertex():
self.hShape.highlightClear()
self.hVertex, self.hShape = None, shape
self.setToolTip(
"Click & drag to move shape '%s'" % shape.label)
self.setStatusTip(self.toolTip())
self.overrideCursor(CURSOR_GRAB)
self.update()
break
else: # Nothing found, clear highlights, reset state.
if self.hShape:
self.hShape.highlightClear()
self.update()
self.hVertex, self.hShape = None, None
self.overrideCursor(CURSOR_DEFAULT)
def mousePressEvent(self, ev):
pos = self.transformPos(ev.pos())
if ev.button() == Qt.LeftButton:
if self.drawing():
# self.handleDrawing(pos) # OLD
if self.current and self.fourpoint: # ADD IF
# Add point to existing shape.
print('Adding points in mousePressEvent is ', self.line[1])
self.current.addPoint(self.line[1])
self.line[0] = self.current[-1]
if self.current.isClosed():
# print('1111')
self.finalise()
elif not self.outOfPixmap(pos):
# Create new shape.
self.current = Shape()# self.current = Shape(shape_type=self.createMode)
self.current.addPoint(pos)
# if self.createMode == "point":
# self.finalise()
# else:
# if self.createMode == "circle":
# self.current.shape_type = "circle"
self.line.points = [pos, pos]
self.setHiding()
self.drawingPolygon.emit(True)
self.update()
else:
selection = self.selectShapePoint(pos)
self.prevPoint = pos
if selection is None:
#pan
QApplication.setOverrideCursor(QCursor(Qt.OpenHandCursor))
self.pan_initial_pos = pos
elif ev.button() == Qt.RightButton and self.editing():
self.selectShapePoint(pos)
self.prevPoint = pos
self.update()
def mouseReleaseEvent(self, ev):
if ev.button() == Qt.RightButton:
menu = self.menus[bool(self.selectedShapeCopy)]
self.restoreCursor()
if not menu.exec_(self.mapToGlobal(ev.pos()))\
and self.selectedShapeCopy:
# Cancel the move by deleting the shadow copy.
self.selectedShapeCopy = None
self.repaint()
elif ev.button() == Qt.LeftButton and self.selectedShape: # OLD
if self.selectedVertex():
self.overrideCursor(CURSOR_POINT)
else:
self.overrideCursor(CURSOR_GRAB)
elif ev.button() == Qt.LeftButton and not self.fourpoint:
pos = self.transformPos(ev.pos())
if self.drawing():
self.handleDrawing(pos)
else:
#pan
QApplication.restoreOverrideCursor() # ?
def endMove(self, copy=False):
assert self.selectedShape and self.selectedShapeCopy
shape = self.selectedShapeCopy
#del shape.fill_color
#del shape.line_color
if copy:
self.shapes.append(shape)
self.selectedShape.selected = False
self.selectedShape = shape
self.repaint()
else:
self.selectedShape.points = [p for p in shape.points]
self.selectedShapeCopy = None
def hideBackroundShapes(self, value):
self.hideBackround = value
if self.selectedShape:
# Only hide other shapes if there is a current selection.
# Otherwise the user will not be able to select a shape.
self.setHiding(True)
self.repaint()
def handleDrawing(self, pos):
if self.current and self.current.reachMaxPoints() is False:
if self.fourpoint:
targetPos = self.line[self.pointnum]
self.current.addPoint(targetPos)
print('current points in handleDrawing is ', self.line[self.pointnum])
self.update()
if self.pointnum == 3:
self.finalise()
else: # 按住送掉后跳到这里
initPos = self.current[0]
print('initPos', self.current[0])
minX = initPos.x()
minY = initPos.y()
targetPos = self.line[1]
maxX = targetPos.x()
maxY = targetPos.y()
self.current.addPoint(QPointF(maxX, minY))
self.current.addPoint(targetPos)
self.current.addPoint(QPointF(minX, maxY))
self.finalise()
elif not self.outOfPixmap(pos):
print('release')
self.current = Shape()
self.current.addPoint(pos)
self.line.points = [pos, pos]
self.setHiding()
self.drawingPolygon.emit(True)
self.update()
def setHiding(self, enable=True):
self._hideBackround = self.hideBackround if enable else False
def canCloseShape(self):
return self.drawing() and self.current and len(self.current) > 2
def mouseDoubleClickEvent(self, ev):
# We need at least 4 points here, since the mousePress handler
# adds an extra one before this handler is called.
if self.canCloseShape() and len(self.current) > 3:
if not self.fourpoint:
self.current.popPoint()
self.finalise()
def selectShape(self, shape):
self.deSelectShape()
shape.selected = True
self.selectedShape = shape
self.setHiding()
self.selectionChanged.emit(True)
self.update()
def selectShapePoint(self, point):
"""Select the first shape created which contains this point."""
self.deSelectShape()
if self.selectedVertex(): # A vertex is marked for selection.
index, shape = self.hVertex, self.hShape
shape.highlightVertex(index, shape.MOVE_VERTEX)
self.selectShape(shape)
return self.hVertex
for shape in reversed(self.shapes):
if self.isVisible(shape) and shape.containsPoint(point):
self.selectShape(shape)
self.calculateOffsets(shape, point)
return self.selectedShape
return None
def calculateOffsets(self, shape, point):
rect = shape.boundingRect()
x1 = rect.x() - point.x()
y1 = rect.y() - point.y()
x2 = (rect.x() + rect.width()) - point.x()
y2 = (rect.y() + rect.height()) - point.y()
self.offsets = QPointF(x1, y1), QPointF(x2, y2)
def snapPointToCanvas(self, x, y):
"""
Moves a point x,y to within the boundaries of the canvas.
:return: (x,y,snapped) where snapped is True if x or y were changed, False if not.
"""
if x < 0 or x > self.pixmap.width() or y < 0 or y > self.pixmap.height():
x = max(x, 0)
y = max(y, 0)
x = min(x, self.pixmap.width())
y = min(y, self.pixmap.height())
return x, y, True
return x, y, False
def boundedMoveVertex(self, pos):
index, shape = self.hVertex, self.hShape
point = shape[index]
if self.outOfPixmap(pos):
size = self.pixmap.size()
clipped_x = min(max(0, pos.x()), size.width())
clipped_y = min(max(0, pos.y()), size.height())
pos = QPointF(clipped_x, clipped_y)
if self.drawSquare:
opposite_point_index = (index + 2) % 4
opposite_point = shape[opposite_point_index]
min_size = min(abs(pos.x() - opposite_point.x()), abs(pos.y() - opposite_point.y()))
directionX = -1 if pos.x() - opposite_point.x() < 0 else 1
directionY = -1 if pos.y() - opposite_point.y() < 0 else 1
shiftPos = QPointF(opposite_point.x() + directionX * min_size - point.x(),
opposite_point.y() + directionY * min_size - point.y())
else:
shiftPos = pos - point
shape.moveVertexBy(index, shiftPos)
lindex = (index + 1) % 4
rindex = (index + 3) % 4
lshift = None
rshift = None
if index % 2 == 0:
rshift = QPointF(shiftPos.x(), 0)
lshift = QPointF(0, shiftPos.y())
else:
lshift = QPointF(shiftPos.x(), 0)
rshift = QPointF(0, shiftPos.y())
shape.moveVertexBy(rindex, rshift)
shape.moveVertexBy(lindex, lshift)
def boundedMoveShape(self, shape, pos):
if self.outOfPixmap(pos):
return False # No need to move
o1 = pos + self.offsets[0]
if self.outOfPixmap(o1):
pos -= QPointF(min(0, o1.x()), min(0, o1.y()))
o2 = pos + self.offsets[1]
if self.outOfPixmap(o2):
pos += QPointF(min(0, self.pixmap.width() - o2.x()),
min(0, self.pixmap.height() - o2.y()))
# The next line tracks the new position of the cursor
# relative to the shape, but also results in making it
# a bit "shaky" when nearing the border and allows it to
# go outside of the shape's area for some reason. XXX
#self.calculateOffsets(self.selectedShape, pos)
dp = pos - self.prevPoint
if dp:
shape.moveBy(dp)
self.prevPoint = pos
return True
return False
def deSelectShape(self):
if self.selectedShape:
self.selectedShape.selected = False
self.selectedShape = None
self.setHiding(False)
self.selectionChanged.emit(False)
self.update()
def deleteSelected(self):
if self.selectedShape:
shape = self.selectedShape
self.shapes.remove(self.selectedShape)
self.selectedShape = None
self.update()
return shape
def copySelectedShape(self):
if self.selectedShape:
shape = self.selectedShape.copy()
self.deSelectShape()
self.shapes.append(shape)
shape.selected = True
self.selectedShape = shape
self.boundedShiftShape(shape)
return shape
def boundedShiftShape(self, shape):
# Try to move in one direction, and if it fails in another.
# Give up if both fail.
point = shape[0]
offset = QPointF(2.0, 2.0)
self.calculateOffsets(shape, point)
self.prevPoint = point
if not self.boundedMoveShape(shape, point - offset):
self.boundedMoveShape(shape, point + offset)
def paintEvent(self, event):
if not self.pixmap:
return super(Canvas, self).paintEvent(event)
p = self._painter
p.begin(self)
p.setRenderHint(QPainter.Antialiasing)
p.setRenderHint(QPainter.HighQualityAntialiasing)
p.setRenderHint(QPainter.SmoothPixmapTransform)
p.scale(self.scale, self.scale)
p.translate(self.offsetToCenter())
p.drawPixmap(0, 0, self.pixmap)
Shape.scale = self.scale
for shape in self.shapes:
if (shape.selected or not self._hideBackround) and self.isVisible(shape):
shape.fill = shape.selected or shape == self.hShape
shape.paint(p)
if self.current:
self.current.paint(p)
self.line.paint(p)
if self.selectedShapeCopy:
self.selectedShapeCopy.paint(p)
# Paint rect
if self.current is not None and len(self.line) == 2 and not self.fourpoint:
# print('Drawing rect')
leftTop = self.line[0]
rightBottom = self.line[1]
rectWidth = rightBottom.x() - leftTop.x()
rectHeight = rightBottom.y() - leftTop.y()
p.setPen(self.drawingRectColor)
brush = QBrush(Qt.BDiagPattern)
p.setBrush(brush)
p.drawRect(leftTop.x(), leftTop.y(), rectWidth, rectHeight)
# ADD:
if (
self.fillDrawing()
and self.fourpoint
and self.current is not None
and len(self.current.points) >= 2
):
print('paint event')
drawing_shape = self.current.copy()
drawing_shape.addPoint(self.line[1])
drawing_shape.fill = True
drawing_shape.paint(p)
if self.drawing() and not self.prevPoint.isNull() and not self.outOfPixmap(self.prevPoint):
p.setPen(QColor(0, 0, 0))
p.drawLine(self.prevPoint.x(), 0, self.prevPoint.x(), self.pixmap.height())
p.drawLine(0, self.prevPoint.y(), self.pixmap.width(), self.prevPoint.y())
self.setAutoFillBackground(True)
if self.verified:
pal = self.palette()
pal.setColor(self.backgroundRole(), QColor(184, 239, 38, 128))
self.setPalette(pal)
else:
pal = self.palette()
pal.setColor(self.backgroundRole(), QColor(232, 232, 232, 255))
self.setPalette(pal)
p.end()
def fillDrawing(self):
return self._fill_drawing
def transformPos(self, point):
"""Convert from widget-logical coordinates to painter-logical coordinates."""
return point / self.scale - self.offsetToCenter()
def offsetToCenter(self):
s = self.scale
area = super(Canvas, self).size()
w, h = self.pixmap.width() * s, self.pixmap.height() * s
aw, ah = area.width(), area.height()
x = (aw - w) / (2 * s) if aw > w else 0
y = (ah - h) / (2 * s) if ah > h else 0
return QPointF(x, y)
def outOfPixmap(self, p):
w, h = self.pixmap.width(), self.pixmap.height()
return not (0 <= p.x() <= w and 0 <= p.y() <= h)
def finalise(self):
assert self.current
if self.current.points[0] == self.current.points[-1]:
# print('finalse')
self.current = None
self.drawingPolygon.emit(False)
self.update()
return
self.current.close()
self.shapes.append(self.current)
self.current = None
self.setHiding(False)
self.newShape.emit()
self.update()
def closeEnough(self, p1, p2):
#d = distance(p1 - p2)
#m = (p1-p2).manhattanLength()
# print "d %.2f, m %d, %.2f" % (d, m, d - m)
return distance(p1 - p2) < self.epsilon
# These two, along with a call to adjustSize are required for the
# scroll area.
def sizeHint(self):
return self.minimumSizeHint()
def minimumSizeHint(self):
if self.pixmap:
return self.scale * self.pixmap.size()
return super(Canvas, self).minimumSizeHint()
def wheelEvent(self, ev):
qt_version = 4 if hasattr(ev, "delta") else 5
if qt_version == 4:
if ev.orientation() == Qt.Vertical:
v_delta = ev.delta()
h_delta = 0
else:
h_delta = ev.delta()
v_delta = 0
else:
delta = ev.angleDelta()
h_delta = delta.x()
v_delta = delta.y()
mods = ev.modifiers()
if Qt.ControlModifier == int(mods) and v_delta:
self.zoomRequest.emit(v_delta)
else:
v_delta and self.scrollRequest.emit(v_delta, Qt.Vertical)
h_delta and self.scrollRequest.emit(h_delta, Qt.Horizontal)
ev.accept()
def keyPressEvent(self, ev):
key = ev.key()
if key == Qt.Key_Escape and self.current:
print('ESC press')
self.current = None
self.drawingPolygon.emit(False)
self.update()
elif key == Qt.Key_Return and self.canCloseShape():
self.finalise()
elif key == Qt.Key_Left and self.selectedShape:
self.moveOnePixel('Left')
elif key == Qt.Key_Right and self.selectedShape:
self.moveOnePixel('Right')
elif key == Qt.Key_Up and self.selectedShape:
self.moveOnePixel('Up')
elif key == Qt.Key_Down and self.selectedShape:
self.moveOnePixel('Down')
def moveOnePixel(self, direction):
# print(self.selectedShape.points)
if direction == 'Left' and not self.moveOutOfBound(QPointF(-1.0, 0)):
# print("move Left one pixel")
self.selectedShape.points[0] += QPointF(-1.0, 0)
self.selectedShape.points[1] += QPointF(-1.0, 0)
self.selectedShape.points[2] += QPointF(-1.0, 0)
self.selectedShape.points[3] += QPointF(-1.0, 0)
elif direction == 'Right' and not self.moveOutOfBound(QPointF(1.0, 0)):
# print("move Right one pixel")
self.selectedShape.points[0] += QPointF(1.0, 0)
self.selectedShape.points[1] += QPointF(1.0, 0)
self.selectedShape.points[2] += QPointF(1.0, 0)
self.selectedShape.points[3] += QPointF(1.0, 0)
elif direction == 'Up' and not self.moveOutOfBound(QPointF(0, -1.0)):
# print("move Up one pixel")
self.selectedShape.points[0] += QPointF(0, -1.0)
self.selectedShape.points[1] += QPointF(0, -1.0)
self.selectedShape.points[2] += QPointF(0, -1.0)
self.selectedShape.points[3] += QPointF(0, -1.0)
elif direction == 'Down' and not self.moveOutOfBound(QPointF(0, 1.0)):
# print("move Down one pixel")
self.selectedShape.points[0] += QPointF(0, 1.0)
self.selectedShape.points[1] += QPointF(0, 1.0)
self.selectedShape.points[2] += QPointF(0, 1.0)
self.selectedShape.points[3] += QPointF(0, 1.0)
self.shapeMoved.emit()
self.repaint()
def moveOutOfBound(self, step):
points = [p1+p2 for p1, p2 in zip(self.selectedShape.points, [step]*4)]
return True in map(self.outOfPixmap, points)
def setLastLabel(self, text, line_color = None, fill_color = None):
assert text
self.shapes[-1].label = text
if line_color:
self.shapes[-1].line_color = line_color
if fill_color:
self.shapes[-1].fill_color = fill_color
return self.shapes[-1]
def undoLastLine(self):
assert self.shapes
self.current = self.shapes.pop()
self.current.setOpen()
self.line.points = [self.current[-1], self.current[0]]
self.drawingPolygon.emit(True)
def resetAllLines(self):
assert self.shapes
self.current = self.shapes.pop()
self.current.setOpen()
self.line.points = [self.current[-1], self.current[0]]
self.drawingPolygon.emit(True)
self.current = None
self.drawingPolygon.emit(False)
self.update()
def loadPixmap(self, pixmap):
self.pixmap = pixmap
self.shapes = []
self.repaint() # 这函数在哪
def loadShapes(self, shapes):
self.shapes = list(shapes)
self.current = None
self.repaint()
def setShapeVisible(self, shape, value):
self.visible[shape] = value
self.repaint()
def currentCursor(self):
cursor = QApplication.overrideCursor()
if cursor is not None:
cursor = cursor.shape()
return cursor
def overrideCursor(self, cursor):
self._cursor = cursor
if self.currentCursor() is None:
QApplication.setOverrideCursor(cursor)
else:
QApplication.changeOverrideCursor(cursor)
def restoreCursor(self):
QApplication.restoreOverrideCursor()
def resetState(self):
self.restoreCursor()
self.pixmap = None
self.update()
def setDrawingShapeToSquare(self, status):
self.drawSquare = status
# Copyright (c) <2015-Present> Tzutalin
# Copyright (C) 2013 MIT, Computer Science and Artificial Intelligence Laboratory. Bryan Russell, Antonio Torralba,
# William T. Freeman. Permission is hereby granted, free of charge, to any person obtaining a copy of this software and
# associated documentation files (the "Software"), to deal in the Software without restriction, including without
# limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the
# Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
# The above copyright notice and this permission notice shall be included in all copies or substantial portions of
# the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT
# NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT
# SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF
# CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
# THE SOFTWARE.
try:
from PyQt5.QtGui import *
from PyQt5.QtCore import *
from PyQt5.QtWidgets import QColorDialog, QDialogButtonBox
except ImportError:
from PyQt4.QtGui import *
from PyQt4.QtCore import *
BB = QDialogButtonBox
class ColorDialog(QColorDialog):
def __init__(self, parent=None):
super(ColorDialog, self).__init__(parent)
self.setOption(QColorDialog.ShowAlphaChannel)
# The Mac native dialog does not support our restore button.
self.setOption(QColorDialog.DontUseNativeDialog)
# Add a restore defaults button.
# The default is set at invocation time, so that it
# works across dialogs for different elements.
self.default = None
self.bb = self.layout().itemAt(1).widget()
self.bb.addButton(BB.RestoreDefaults)
self.bb.clicked.connect(self.checkRestore)
def getColor(self, value=None, title=None, default=None):
self.default = default
if title:
self.setWindowTitle(title)
if value:
self.setCurrentColor(value)
return self.currentColor() if self.exec_() else None
def checkRestore(self, button):
if self.bb.buttonRole(button) & BB.ResetRole and self.default:
self.setCurrentColor(self.default)
# Copyright (c) <2015-Present> Tzutalin
# Copyright (C) 2013 MIT, Computer Science and Artificial Intelligence Laboratory. Bryan Russell, Antonio Torralba,
# William T. Freeman. Permission is hereby granted, free of charge, to any person obtaining a copy of this software and
# associated documentation files (the "Software"), to deal in the Software without restriction, including without
# limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the
# Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
# The above copyright notice and this permission notice shall be included in all copies or substantial portions of
# the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT
# NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT
# SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF
# CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
# THE SOFTWARE.
SETTING_FILENAME = 'filename'
SETTING_RECENT_FILES = 'recentFiles'
SETTING_WIN_SIZE = 'window/size'
SETTING_WIN_POSE = 'window/position'
SETTING_WIN_GEOMETRY = 'window/geometry'
SETTING_LINE_COLOR = 'line/color'
SETTING_FILL_COLOR = 'fill/color'
SETTING_ADVANCE_MODE = 'advanced'
SETTING_WIN_STATE = 'window/state'
SETTING_SAVE_DIR = 'savedir'
SETTING_PAINT_LABEL = 'paintlabel'
SETTING_LAST_OPEN_DIR = 'lastOpenDir'
SETTING_AUTO_SAVE = 'autosave'
SETTING_SINGLE_CLASS = 'singleclass'
FORMAT_PASCALVOC='PascalVOC'
FORMAT_YOLO='YOLO'
SETTING_DRAW_SQUARE = 'draw/square'
SETTING_LABEL_FILE_FORMAT= 'labelFileFormat'
DEFAULT_ENCODING = 'utf-8'
# Copyright (c) <2015-Present> Tzutalin
# Copyright (C) 2013 MIT, Computer Science and Artificial Intelligence Laboratory. Bryan Russell, Antonio Torralba,
# William T. Freeman. Permission is hereby granted, free of charge, to any person obtaining a copy of this software and
# associated documentation files (the "Software"), to deal in the Software without restriction, including without
# limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the
# Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
# The above copyright notice and this permission notice shall be included in all copies or substantial portions of
# the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT
# NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT
# SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF
# CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
# THE SOFTWARE.
#!/usr/bin/env python
# -*- coding: utf8 -*-
import json
from pathlib import Path
from libs.constants import DEFAULT_ENCODING
import os
JSON_EXT = '.json'
ENCODE_METHOD = DEFAULT_ENCODING
class CreateMLWriter:
def __init__(self, foldername, filename, imgsize, shapes, outputfile, databasesrc='Unknown', localimgpath=None):
self.foldername = foldername
self.filename = filename
self.databasesrc = databasesrc
self.imgsize = imgsize
self.boxlist = []
self.localimgpath = localimgpath
self.verified = False
self.shapes = shapes
self.outputfile = outputfile
def write(self):
if os.path.isfile(self.outputfile):
with open(self.outputfile, "r") as file:
input_data = file.read()
outputdict = json.loads(input_data)
else:
outputdict = []
outputimagedict = {
"image": self.filename,
"annotations": []
}
for shape in self.shapes:
points = shape["points"]
x1 = points[0][0]
y1 = points[0][1]
x2 = points[1][0]
y2 = points[2][1]
height, width, x, y = self.calculate_coordinates(x1, x2, y1, y2)
shapedict = {
"label": shape["label"],
"coordinates": {
"x": x,
"y": y,
"width": width,
"height": height
}
}
outputimagedict["annotations"].append(shapedict)
# check if image already in output
exists = False
for i in range(0, len(outputdict)):
if outputdict[i]["image"] == outputimagedict["image"]:
exists = True
outputdict[i] = outputimagedict
break
if not exists:
outputdict.append(outputimagedict)
Path(self.outputfile).write_text(json.dumps(outputdict), ENCODE_METHOD)
def calculate_coordinates(self, x1, x2, y1, y2):
if x1 < x2:
xmin = x1
xmax = x2
else:
xmin = x2
xmax = x1
if y1 < y2:
ymin = y1
ymax = y2
else:
ymin = y2
ymax = y1
width = xmax - xmin
if width < 0:
width = width * -1
height = ymax - ymin
# x and y from center of rect
x = xmin + width / 2
y = ymin + height / 2
return height, width, x, y
class CreateMLReader:
def __init__(self, jsonpath, filepath):
self.jsonpath = jsonpath
self.shapes = []
self.verified = False
self.filename = filepath.split("/")[-1:][0]
try:
self.parse_json()
except ValueError:
print("JSON decoding failed")
def parse_json(self):
with open(self.jsonpath, "r") as file:
inputdata = file.read()
outputdict = json.loads(inputdata)
self.verified = True
if len(self.shapes) > 0:
self.shapes = []
for image in outputdict:
if image["image"] == self.filename:
for shape in image["annotations"]:
self.add_shape(shape["label"], shape["coordinates"])
def add_shape(self, label, bndbox):
xmin = bndbox["x"] - (bndbox["width"] / 2)
ymin = bndbox["y"] - (bndbox["height"] / 2)
xmax = bndbox["x"] + (bndbox["width"] / 2)
ymax = bndbox["y"] + (bndbox["height"] / 2)
points = [(xmin, ymin), (xmax, ymin), (xmax, ymax), (xmin, ymax)]
self.shapes.append((label, points, None, None, True))
def get_shapes(self):
return self.shapes
# Copyright (c) <2015-Present> Tzutalin
# Copyright (C) 2013 MIT, Computer Science and Artificial Intelligence Laboratory. Bryan Russell, Antonio Torralba,
# William T. Freeman. Permission is hereby granted, free of charge, to any person obtaining a copy of this software and
# associated documentation files (the "Software"), to deal in the Software without restriction, including without
# limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the
# Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
# The above copyright notice and this permission notice shall be included in all copies or substantial portions of
# the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT
# NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT
# SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF
# CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
# THE SOFTWARE.
#!/usr/bin/env python
# -*- coding: utf-8 -*-
import sys
try:
from PyQt5.QtGui import *
from PyQt5.QtCore import *
from PyQt5.QtWidgets import *
except ImportError:
# needed for py3+qt4
# Ref:
# http://pyqt.sourceforge.net/Docs/PyQt4/incompatible_apis.html
# http://stackoverflow.com/questions/21217399/pyqt4-qtcore-qvariant-object-instead-of-a-string
if sys.version_info.major >= 3:
import sip
sip.setapi('QVariant', 2)
from PyQt4.QtGui import *
from PyQt4.QtCore import *
# PyQt5: TypeError: unhashable type: 'QListWidgetItem'
class HashableQListWidgetItem(QListWidgetItem):
def __init__(self, *args):
super(HashableQListWidgetItem, self).__init__(*args)
def __hash__(self):
return hash(id(self))
# Copyright (c) <2015-Present> Tzutalin
# Copyright (C) 2013 MIT, Computer Science and Artificial Intelligence Laboratory. Bryan Russell, Antonio Torralba,
# William T. Freeman. Permission is hereby granted, free of charge, to any person obtaining a copy of this software and
# associated documentation files (the "Software"), to deal in the Software without restriction, including without
# limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the
# Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
# The above copyright notice and this permission notice shall be included in all copies or substantial portions of
# the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT
# NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT
# SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF
# CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
# THE SOFTWARE.
try:
from PyQt5.QtGui import *
from PyQt5.QtCore import *
from PyQt5.QtWidgets import *
except ImportError:
from PyQt4.QtGui import *
from PyQt4.QtCore import *
from libs.utils import newIcon, labelValidator
BB = QDialogButtonBox
class LabelDialog(QDialog):
def __init__(self, text="Enter object label", parent=None, listItem=None):
super(LabelDialog, self).__init__(parent)
self.edit = QLineEdit() # OLD
# self.edit = QTextEdit()
self.edit.setText(text)
# self.edit.setValidator(labelValidator()) # 验证有效性
self.edit.editingFinished.connect(self.postProcess)
model = QStringListModel()
model.setStringList(listItem)
completer = QCompleter()
completer.setModel(model)
self.edit.setCompleter(completer)
layout = QVBoxLayout()
layout.addWidget(self.edit)
self.buttonBox = bb = BB(BB.Ok | BB.Cancel, Qt.Horizontal, self)
bb.button(BB.Ok).setIcon(newIcon('done'))
bb.button(BB.Cancel).setIcon(newIcon('undo'))
bb.accepted.connect(self.validate)
bb.rejected.connect(self.reject)
layout.addWidget(bb)
# if listItem is not None and len(listItem) > 0:
# self.listWidget = QListWidget(self)
# for item in listItem:
# self.listWidget.addItem(item)
# self.listWidget.itemClicked.connect(self.listItemClick)
# self.listWidget.itemDoubleClicked.connect(self.listItemDoubleClick)
# layout.addWidget(self.listWidget)
self.setLayout(layout)
def validate(self):
try:
if self.edit.text().trimmed():
self.accept()
except AttributeError:
# PyQt5: AttributeError: 'str' object has no attribute 'trimmed'
if self.edit.text().strip():
self.accept()
def postProcess(self):
try:
self.edit.setText(self.edit.text().trimmed())
# print(self.edit.text())
except AttributeError:
# PyQt5: AttributeError: 'str' object has no attribute 'trimmed'
self.edit.setText(self.edit.text())
print(self.edit.text())
def popUp(self, text='', move=True):
self.edit.setText(text)
self.edit.setSelection(0, len(text))
self.edit.setFocus(Qt.PopupFocusReason)
if move:
cursor_pos = QCursor.pos()
parent_bottomRight = self.parentWidget().geometry()
max_x = parent_bottomRight.x() + parent_bottomRight.width() - self.sizeHint().width()
max_y = parent_bottomRight.y() + parent_bottomRight.height() - self.sizeHint().height()
max_global = self.parentWidget().mapToGlobal(QPoint(max_x, max_y))
if cursor_pos.x() > max_global.x():
cursor_pos.setX(max_global.x())
if cursor_pos.y() > max_global.y():
cursor_pos.setY(max_global.y())
self.move(cursor_pos)
return self.edit.text() if self.exec_() else None
def listItemClick(self, tQListWidgetItem):
try:
text = tQListWidgetItem.text().trimmed()
except AttributeError:
# PyQt5: AttributeError: 'str' object has no attribute 'trimmed'
text = tQListWidgetItem.text().strip()
self.edit.setText(text)
def listItemDoubleClick(self, tQListWidgetItem):
self.listItemClick(tQListWidgetItem)
self.validate()
# Copyright (c) 2016 Tzutalin
# Create by TzuTaLin <tzu.ta.lin@gmail.com>
try:
from PyQt5.QtGui import QImage
except ImportError:
from PyQt4.QtGui import QImage
from base64 import b64encode, b64decode
from libs.pascal_voc_io import PascalVocWriter
from libs.yolo_io import YOLOWriter
from libs.pascal_voc_io import XML_EXT
from enum import Enum
import os.path
import sys
class LabelFileFormat(Enum):
PASCAL_VOC= 1
YOLO = 2
class LabelFileError(Exception):
pass
class LabelFile(object):
# It might be changed as window creates. By default, using XML ext
# suffix = '.lif'
suffix = XML_EXT
def __init__(self, filename=None):
self.shapes = ()
self.imagePath = None
self.imageData = None
self.verified = False
def savePascalVocFormat(self, filename, shapes, imagePath, imageData,
lineColor=None, fillColor=None, databaseSrc=None):
imgFolderPath = os.path.dirname(imagePath)
imgFolderName = os.path.split(imgFolderPath)[-1]
imgFileName = os.path.basename(imagePath)
#imgFileNameWithoutExt = os.path.splitext(imgFileName)[0]
# Read from file path because self.imageData might be empty if saving to
# Pascal format
image = QImage()
image.load(imagePath)
imageShape = [image.height(), image.width(),
1 if image.isGrayscale() else 3]
writer = PascalVocWriter(imgFolderName, imgFileName,
imageShape, localImgPath=imagePath)
writer.verified = self.verified
for shape in shapes:
points = shape['points']
label = shape['label']
# Add Chris
difficult = int(shape['difficult'])
bndbox = LabelFile.convertPoints2BndBox(points)
writer.addBndBox(bndbox[0], bndbox[1], bndbox[2], bndbox[3], label, difficult)
writer.save(targetFile=filename)
return
def saveYoloFormat(self, filename, shapes, imagePath, imageData, classList,
lineColor=None, fillColor=None, databaseSrc=None):
imgFolderPath = os.path.dirname(imagePath)
imgFolderName = os.path.split(imgFolderPath)[-1]
imgFileName = os.path.basename(imagePath)
#imgFileNameWithoutExt = os.path.splitext(imgFileName)[0]
# Read from file path because self.imageData might be empty if saving to
# Pascal format
image = QImage()
image.load(imagePath)
imageShape = [image.height(), image.width(),
1 if image.isGrayscale() else 3]
writer = YOLOWriter(imgFolderName, imgFileName,
imageShape, localImgPath=imagePath)
writer.verified = self.verified
for shape in shapes:
points = shape['points']
label = shape['label']
# Add Chris
difficult = int(shape['difficult'])
bndbox = LabelFile.convertPoints2BndBox(points)
writer.addBndBox(bndbox[0], bndbox[1], bndbox[2], bndbox[3], label, difficult)
writer.save(targetFile=filename, classList=classList)
return
def toggleVerify(self):
self.verified = not self.verified
''' ttf is disable
def load(self, filename):
import json
with open(filename, 'rb') as f:
data = json.load(f)
imagePath = data['imagePath']
imageData = b64decode(data['imageData'])
lineColor = data['lineColor']
fillColor = data['fillColor']
shapes = ((s['label'], s['points'], s['line_color'], s['fill_color'])\
for s in data['shapes'])
# Only replace data after everything is loaded.
self.shapes = shapes
self.imagePath = imagePath
self.imageData = imageData
self.lineColor = lineColor
self.fillColor = fillColor
def save(self, filename, shapes, imagePath, imageData, lineColor=None, fillColor=None):
import json
with open(filename, 'wb') as f:
json.dump(dict(
shapes=shapes,
lineColor=lineColor, fillColor=fillColor,
imagePath=imagePath,
imageData=b64encode(imageData)),
f, ensure_ascii=True, indent=2)
'''
@staticmethod
def isLabelFile(filename):
fileSuffix = os.path.splitext(filename)[1].lower()
return fileSuffix == LabelFile.suffix
@staticmethod
def convertPoints2BndBox(points):
xmin = float('inf')
ymin = float('inf')
xmax = float('-inf')
ymax = float('-inf')
for p in points:
x = p[0]
y = p[1]
xmin = min(x, xmin)
ymin = min(y, ymin)
xmax = max(x, xmax)
ymax = max(y, ymax)
# Martin Kersner, 2015/11/12
# 0-valued coordinates of BB caused an error while
# training faster-rcnn object detector.
if xmin < 1:
xmin = 1
if ymin < 1:
ymin = 1
return (int(xmin), int(ymin), int(xmax), int(ymax))
# Copyright (c) <2015-Present> Tzutalin
# Copyright (C) 2013 MIT, Computer Science and Artificial Intelligence Laboratory. Bryan Russell, Antonio Torralba,
# William T. Freeman. Permission is hereby granted, free of charge, to any person obtaining a copy of this software and
# associated documentation files (the "Software"), to deal in the Software without restriction, including without
# limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the
# Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
# The above copyright notice and this permission notice shall be included in all copies or substantial portions of
# the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT
# NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT
# SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF
# CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
# THE SOFTWARE.
#!/usr/bin/env python
# -*- coding: utf8 -*-
import sys
from xml.etree import ElementTree
from xml.etree.ElementTree import Element, SubElement
from lxml import etree
import codecs
from libs.constants import DEFAULT_ENCODING
from libs.ustr import ustr
XML_EXT = '.xml'
ENCODE_METHOD = DEFAULT_ENCODING
class PascalVocWriter:
def __init__(self, foldername, filename, imgSize,databaseSrc='Unknown', localImgPath=None):
self.foldername = foldername
self.filename = filename
self.databaseSrc = databaseSrc
self.imgSize = imgSize
self.boxlist = []
self.localImgPath = localImgPath
self.verified = False
def prettify(self, elem):
"""
Return a pretty-printed XML string for the Element.
"""
rough_string = ElementTree.tostring(elem, 'utf8')
root = etree.fromstring(rough_string)
return etree.tostring(root, pretty_print=True, encoding=ENCODE_METHOD).replace(" ".encode(), "\t".encode())
# minidom does not support UTF-8
'''reparsed = minidom.parseString(rough_string)
return reparsed.toprettyxml(indent="\t", encoding=ENCODE_METHOD)'''
def genXML(self):
"""
Return XML root
"""
# Check conditions
if self.filename is None or \
self.foldername is None or \
self.imgSize is None:
return None
top = Element('annotation')
if self.verified:
top.set('verified', 'yes')
folder = SubElement(top, 'folder')
folder.text = self.foldername
filename = SubElement(top, 'filename')
filename.text = self.filename
if self.localImgPath is not None:
localImgPath = SubElement(top, 'path')
localImgPath.text = self.localImgPath
source = SubElement(top, 'source')
database = SubElement(source, 'database')
database.text = self.databaseSrc
size_part = SubElement(top, 'size')
width = SubElement(size_part, 'width')
height = SubElement(size_part, 'height')
depth = SubElement(size_part, 'depth')
width.text = str(self.imgSize[1])
height.text = str(self.imgSize[0])
if len(self.imgSize) == 3:
depth.text = str(self.imgSize[2])
else:
depth.text = '1'
segmented = SubElement(top, 'segmented')
segmented.text = '0'
return top
def addBndBox(self, xmin, ymin, xmax, ymax, name, difficult):
bndbox = {'xmin': xmin, 'ymin': ymin, 'xmax': xmax, 'ymax': ymax}
bndbox['name'] = name
bndbox['difficult'] = difficult
self.boxlist.append(bndbox)
def appendObjects(self, top):
for each_object in self.boxlist:
object_item = SubElement(top, 'object')
name = SubElement(object_item, 'name')
name.text = ustr(each_object['name'])
pose = SubElement(object_item, 'pose')
pose.text = "Unspecified"
truncated = SubElement(object_item, 'truncated')
if int(float(each_object['ymax'])) == int(float(self.imgSize[0])) or (int(float(each_object['ymin']))== 1):
truncated.text = "1" # max == height or min
elif (int(float(each_object['xmax']))==int(float(self.imgSize[1]))) or (int(float(each_object['xmin']))== 1):
truncated.text = "1" # max == width or min
else:
truncated.text = "0"
difficult = SubElement(object_item, 'difficult')
difficult.text = str( bool(each_object['difficult']) & 1 )
bndbox = SubElement(object_item, 'bndbox')
xmin = SubElement(bndbox, 'xmin')
xmin.text = str(each_object['xmin'])
ymin = SubElement(bndbox, 'ymin')
ymin.text = str(each_object['ymin'])
xmax = SubElement(bndbox, 'xmax')
xmax.text = str(each_object['xmax'])
ymax = SubElement(bndbox, 'ymax')
ymax.text = str(each_object['ymax'])
def save(self, targetFile=None):
root = self.genXML()
self.appendObjects(root)
out_file = None
if targetFile is None:
out_file = codecs.open(
self.filename + XML_EXT, 'w', encoding=ENCODE_METHOD)
else:
out_file = codecs.open(targetFile, 'w', encoding=ENCODE_METHOD)
prettifyResult = self.prettify(root)
out_file.write(prettifyResult.decode('utf8'))
out_file.close()
class PascalVocReader:
def __init__(self, filepath):
# shapes type:
# [labbel, [(x1,y1), (x2,y2), (x3,y3), (x4,y4)], color, color, difficult]
self.shapes = []
self.filepath = filepath
self.verified = False
try:
self.parseXML()
except:
pass
def getShapes(self):
return self.shapes
def addShape(self, label, bndbox, difficult):
xmin = int(float(bndbox.find('xmin').text))
ymin = int(float(bndbox.find('ymin').text))
xmax = int(float(bndbox.find('xmax').text))
ymax = int(float(bndbox.find('ymax').text))
points = [(xmin, ymin), (xmax, ymin), (xmax, ymax), (xmin, ymax)]
self.shapes.append((label, points, None, None, difficult))
def parseXML(self):
assert self.filepath.endswith(XML_EXT), "Unsupport file format"
parser = etree.XMLParser(encoding=ENCODE_METHOD)
xmltree = ElementTree.parse(self.filepath, parser=parser).getroot()
filename = xmltree.find('filename').text
try:
verified = xmltree.attrib['verified']
if verified == 'yes':
self.verified = True
except KeyError:
self.verified = False
for object_iter in xmltree.findall('object'):
bndbox = object_iter.find("bndbox")
label = object_iter.find('name').text
# Add chris
difficult = False
if object_iter.find('difficult') is not None:
difficult = bool(int(object_iter.find('difficult').text))
self.addShape(label, bndbox, difficult)
return True
This source diff could not be displayed because it is too large. You can view the blob instead.
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment